Relevance Of The uMap Collaborative Platform As A Support For Choropleth Mapping: An Atlas Applied To All-Cause Excess Mortality Alerts By Traffic Light - 1st French Containment In 2020

Relevance Of The uMap Collaborative Platform As A Support For Choropleth Mapping: An Atlas Applied To All-Cause Excess Mortality Alerts By Traffic Light - 1st French Containment In 2020 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Relevance Of The uMap Collaborative Platform As A Support For Choropleth Mapping: An Atlas Applied To All-Cause Excess Mortality Alerts By Traffic Light - 1st French Containment In 2020 Anne QUESNEL-BARBET, Thierry PAGES, Julien SOULA, Gilles MAIGNANT, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4796017/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In the wake of the health crisis, for the year 2020 we have created a France-wide geomatic project to produce several atlases of mortality for all pathologies, with very fine grids at commune and arrondissement levels (Marseille, Lyon, and Paris). The aim is to bring to the collaborative map-sharing platform uMap environment, original visualization and knowledge, and decision-making aids, complementary to existing information and relevant to both the general public and healthcare professionals. Method: We followed a two-step geomatic action plan (with and without webmapping) to create, from each of our four variables, an atlas of the 1 st French containment period. Interpretation of the atlas is facilitated by the graphical display of colored choropleth maps with legend and statistical tooltip, and by rapid transmission to the user of an "Excess Mortality alert". In two different webmapping environments tested in advance, our results only display uMap and its TrafficLight atlas, that instantly transmits the Excess Mortality alert by means of semantic interplay of colors. Comparison between regions is facilitated by the display of administrative contours and by a layer and legend management tool. Today, uMap delivers the expected results without having to worry about its remote server managed by OpenStreetMap, and its interoperability allows atlases to be exported and printed. Outlook : Having taken a step back from geomatic technologies, uMap shows great potential for its implementation dedicated to healthcare on a local server. Further development, to create and associate legends with choropleth maps, could increase uMap’s functionality and interest. Conclusions: We have innovated, demonstrated and reinforced the value of the collaborative and interoperable uMap mapping platform for visual rendering by instantiating our thematic atlases with choropleth maps and their legends. The TrafficLight atlas is highly relevant for the instant message it communicates by commune, and in this way, we have provided additional information easy to interpretable for all audiences. Finally, uMap offers many advantages in terms of licensing, design, and use of the atlases. This encourages us to continue improving it with the help of its contributors, and to explore it further, with an optimized geomatic action plan. Geomatic Geographic Mapping Choropleth Mapping Collaborative Atlas Platform Mortality Spatial Analysis Alert System Digital Health Medical Informatics. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 1. Introduction The needs and interests in geomatics of the medical sector, particularly hospitals, have led us over time to strengthen our interdisciplinary collaborations on joint research projects. In response to the health crisis, our new health geomatics project led us to create a mortality atlas targeting "all pathologies that can be visualized on very fine grids of the communes and arrondissements of Marseille, Lyon and Paris on the scale of France". Our monitoring of geomatics technologies, IT development, and data sources has led us to focus on two atlas environments. Our main three-point objective is firstly to provide an original and relevant decision-making aid in a geomatics environment (on a French scale), in the form of a cartographic atlas whose information is processed at a fine territorial grid (communes and arrondissements). This processed information then provides new insights compared with studies whose information is aggregated at coarser departmental and regional levels, leading to visualization biases e.g. a mesh that is too aggregated can hide, render invisible or imperceptible other information remaining to be discovered [1]. The second point is to process information on mortality in 2020 (compared with 2018 and 2019) based on the period of the first French containment. Finally, the third point is to make the thematic information mapped on the scale of France and its ( Départements et Régions d'Outre-Mer (DROMs/OSDRs, the OverSeas Departments and Regions) easily understandable to the general public and users of our atlases in the uMap environment . To this end, we have constructed atlas-based statistical indicators of excess mortality. These over-under mortality indicators (per 10,000 inhabitants at municipal level) are also displayed in the tooltips. Users can connect to the uMap environment for each atlas instance associated with a statistical variable, in four layers representing four major regions: northeast, northwest, southeast and southwest. The main methodological approaches for obtaining results are divided into two stages: "step 1, geomatics without webmapping" and "step 2, geomatics with webmapping" [2]. Our atlas instances (uMap showcases) are the main results of our study. Among the three e-atlases, the " FeuTricolore " (TrafficLight) thematic atlas created from the “Khi2LnFr” variable, displays one of three colors: red, orange, and green, if the commune is exclusively concerned by an excess mortality alert. The latter atlas appears even more relevant, as cartographic information is instantly visible and easily interpreted by the human eye. We have qualified this "FeuTricolore" (TrafficLight) atlas as the best messenger. After a review of the existing geomatics environment (history of digital cartography, its standards and health geomatics), this article describes the objective reasons for using the uMap geomatics environment as a support for thematic health atlases (composed of choropleth maps). It describes the methods and processes involved in developing the uMap environment, followed by results on uMap usability (platform ergonomics) and cartography (information visualization and interpretation). Now, we present discussions on the geomatic process (uMap software package), and on the advantages and disadvantages in terms of design and use. The outlook then looks at how the uMap environment and the tasks can be assessed in terms of its performance, attractiveness, and profitability, and what new developments, geomatic instantiations and other improvements to the environment can be envisaged. We end with our conclusions. 2. Background 2.1. State of the art in geomatics, healthcare geomatics 2.1.1. History of digital cartography Geography is an integral part of our modern digital world, and the map is its main tool. The map is a model of partial representation of the geographic space under analysis; many different types of maps are produced, with varying degrees of respect for scientific reasoning and graphic semiology standards [3-5]. Nowadays, maps are most often produced using Cartographie et Dessin Assistés par Ordinateur ( CAO-DAO/CAMD Computer Aided Mapping and Drawing) software, or from software environments with multiple functionalities for spatial representation and analysis, such as Systèmes d’Information Géographique ( SIGs/GISs, Geographic Information Systems) and shared geographic information platforms: all these environments represent the world of geomatics. Geomatics is a contraction of the words "geography" and "computer science» and is associated with the world of digital cartography and spatial analysis. According to Henri Pornon, it is the discipline that deals with the management of spatially referenced data, involving the sciences and technologies associated with their acquisition, storage, processing, and dissemination. A distinction is made between "geographic data", representative of an object in the territory, and "localized data", representative of an object positioned in the territory [6]. By contraction of geographic and localized, we speak of geolocalized data for positioned object. In other words, geomatics is a group of IT methods and tools designed to collect, organize, analyze, and represent - in the sense of modeling - geolocalized data. It enables geolocalized information to beacquired, structured, integrated and analyzed, before the results are published in the form of maps. Geomatics is generally associated with SIG/GIS as the main digital tool implementing the concepts and methods of geography [6]. Geomatics has no existence without geolocalized data, which has been evolving since the 1970s, with the production, distribution, and use of public geolocalized data[6]. The recent concept of Trouvable, Accessible, Interopérable, Réutilisable données ( FAIR, Findable, Accessible, Interoperable, Reusable data) [7, 8] is defined as the ability for IT systems (with little or no human intervention) to manage information, in compliance with the four rules: Findability, Accessibility, Interoperability and Reusability. The quality of data production is also under scientific scrutiny in the collaborative Projet OpenStreetMap, carte du monde modifiable gratuitement ( OSM, OpenStreetMap project: a free editable map of the world) [9-11]. Numerous concepts have also emerged from this ongoing development of rigorous data management, that impacts geolocalized data by inheritance. We refer to open data - free databases[1] [12, 13], crowdsourcing and crowd mapping - participative production and mapping [14-16]; to open-source - free and open computer code with authorized modification and redistribution; to open source model - provision of software and open/free code as opposed to proprietary code [17, 18]; or to participatory Web 2.0 or contributory internet - shared information and geographic information platform; collaborative platform; information commons and commons [19]. Location-based information and free geomatics[2] [9] are by heritage affected by the "4 rules of FAIR data, Findable, Accessible, Interoperable, Reusable" of free computing [7, 8]. These include interoperability, which guarantees (thanks to open standards[3]) the exchange of files between users equipped with different hardware or software [20]. Two types of interoperability can be distinguished: syntactic interoperability, allowing two systems to communicate with one another, and inter-domain interoperability, allowing several organizations to cooperate and exchange information with one another [21]. Interoperability thus offers proprietary producers the opportunity to turn to these concepts, gradually leading to the emergence of production chains and pipelines[4] that are completely free to use [9, 22, 23]. Three categories of open standards accompany interoperability[24]. (1)- standards set by software publishers and users (e.g., the "shapefile .shp " file format from ©ESRI, a SIG/GIS publisher, or the "DraWinG .dwg " file format from ©AutoCAD, a drawing software publisher). (2)- formal standardsdefined according to a strict protocol by standards bodies such as the Consortium Géospatial Ouvert ( OGC , Open Geospatial Consortium ( OGC [5], 1994)) [7] which represents a worldwide resource community for geospatial information and standards. Other representative bodies include the Association Française de NORmalisation ( AFNOR/FSA , French Standards Association) [25], Organisation Internationale de Normalisation (ISO , International Organization for Standardization) [26], the Comité Européen de Normalisation ( CEN/ECS , European Committee for Standardization) and Etalab [27] of the Département de la Direction Interministérielle du NUMérique ( DINUM/IDD , Interministerial Digital Department), whose aim is to improve public service and public action through data; (3)- French government standards created to adapt uses to the French national territory. Thus, the Conseil National de l'Information Géolocalisée ( CNIG/NCGI , National Council for Geolocated Information) - newly renamed ( CNIG for geographic becomes CNIG for Geolocalized/Geolocated ) to cover the topography of territories and, more recently, their location - coordinates and supports public and private players in meeting the evolving challenges of standardization, innovation, production, distribution and sharing of geolocated information. Finally, standards bodies such as the Consortium Géospatial Ouvert ( OGC, Open Geospatial Consortium) and the Conseil National de l'Information Géolocalisée ( CNIG/NCGI, National Council for Geolocated Information ) comply with European policies, directives and regulations: Infrastructure d’Information Spatiale en Europe ( INSPIRE Directive , Infrastructure for Spatial Information in Europe), the European fair and innovative data (DataAct) [28]. The following is a selection of major projects and standards impacting geomatics: 1. The standards of the international association of Producteurs de Pétrole et de Gaz ( PPG/OGP , Oil and Gas Producers), formerly Groupe Européen d’Etude du Pétrole ( EPSG , European Petroleum Survey Group) 1985 for geodesy maintenance and free sharing of the EPSG geodetic parameter dataset [29-32]. 2. Web standards[6], including the World Wide Web Consortium ( W3C ) for developers and by heritage for geomatics developers [13, 24, 33]. 3. National projects based on very large-scale building information modeling standards (BIM and GeoBIM) [34, 35]. 4. The Groupe de Travail sur l’Information Géospatiale de la Défense ( GTIGD/DGIWG, The Defence Geospatial Information Working Group) has been managing geospatial information and ensuring interoperability since 1983 by creating the standards, implementation guidelines and procedures necessary to enable the provision, exchange and use of standardized geospatial information [36]. 2.1.2. Geomatics in healthcare - themes, visualization, territorial networking As seen above, geomatics is benefiting from a continuous digital transformation. The map is at the center of attention in many fields of business and research. Open-source geomatics is accelerating this trend, meaning that cartography is no longer the sole preserve of geolocation information specialists. Many professionals, such as IT specialists, data scientists, open-source project contributors, etc., are now involved in geomatics [9] and scientists (whether or not they specialize in location-based information) have seized on the map as a pretty, useful, strategic, and valuable object; an object supporting knowledge, spatial memorization and reflection and/or as a digital object to be produced, automated and distributed. Geomatics in healthcare has become increasingly important in recent decades. Current societal developments, marked by the covid 19 pandemic, have boosted the use of digital maps in healthcare. At the height of the crisis, maps were followed by millions of Internet users and other readers. This societal awareness of health risks, accentuated by the covid pandemic, calls forthe map to become an essential part of our societies and Public Health. New map showcases - based on geomatics technologies - are emerging in the form of online dashboards, replacing more complex SIG/GIS environments. These dashboards can be public, private or mixed (the private part is accessible with authorization), and offer users visualization and interaction functionalities [37, 38]. Static or dynamic atlases are integrated into these environments, with geographical scales ranging from global to local, using geolocalized health data that are more or less aggregated according to the authorized territorial mesh for spatio-statistical analysis, which can lead to a number of biases [1]. 2.1.2.1. Cartographic information bias - four causes of over-aggregated meshes The use of meshes that are too aggregated in cartography can lead to information bias for four reasons identified by the authors [39]. (1) A loss of accuracy can occur when using a mesh that is too large. (2) The "aggregation problem" arises if the analysis is misinterpreted due to data appearing more homogeneous than they actually are. (3) There may be difficulties in visualizing inter-zone disparities or geographical variations. (4) Finally, a phenomenon of exaggeration or over-representation of information for sparsely populated areas may appear due to an overly wide mesh wrongly accentuating the sanitary phenomenon [40]. As a result, maps with mesh sizes that are too aggregated can lead to errors of interpretation and planning, and thus to ineffective public health interventions, for example. It is therefore important to use maps with the most appropriate mesh size possible, in order to address the limitations outlined above [41]. 2.2. Interests and needs identified by the hospital sector for geomatics in healthcare In the literature, scientific publications in the healthcare sector bear witness to the growing interest in geomatics; this interest, accentuated by the covid health crisis, was already focused before the pandemic on various medical and surgical disciplines, or on public health themes such as assistance with prevention and healthcare organization [6, 42-44]. 2.2.1. Point of view the hospital group of the Institut Catholique de Lille Geomatics offers numerous advantages for representing health indicators in the hospital sector, particularly in evaluating epidemiological surveillance and assessing patient retention or outflow, alongside recent developments in the medico-economic aspects of care pathways. Geomatics indeed helps understanding morbidity and mortality indicators and addressing public health needs within a specific geographic area based on population responsibility. Epidemiological monitoring offers several advantages for the Groupement des Hôpitaux de l'Institut Catholique de Lille (GHICL ,Group of hospitals of the Institut Catholique de Lille) · It enables better adjustment of care production strategies and improves capacity planning for patient care, concretely demonstrated by the creation of bed-manager positions. This allows the healthcare establishment to optimize resource allocation for various treatments—medical, surgical, or obstetrical—by adjusting the number of hospital beds and the necessary medical and nursing staff. It enhances logistical management of home care, including the optimization of scheduling for healthcare professionals attending to patients receiving home hospitalization. · Additionally, geomatics can be utilized to assess the quality of care by analyzing geospatial data. This is crucial for monitoring epidemics and overseeing patient care, whether by the originating facility or other nearby healthcare structures. 2.2.2. State of the art in mortality atlases Generally speaking, epidemiological statistics and the mortality atlases they produce are essential tools for understanding mortality patterns and trends in a given region, zone or country and for comparing the mortality rates of the studied areas [45]. In France,for example, the Institut National de la Statistique et des Études Économiques (INSEE , the French National Institute of Statistics and Economic Studies), systematically publishes an interpretation of the country's mortality data, as well as several mortality atlases at sub-national (regional or departmental) level. Overall, these atlases show mortality rates by cause of death, age and sex. They provide a better understanding of geographical variations and identify the most vulnerable population groups [46]. Two French institutions stand out: Santé Publique France (SPF/FPH France Public Health) (formerly INstitut de Veille Sanitaire-INVS ) and its “Géodes” platform [47]. SPF/FPH is an organization in charge of continuous monitoring of the population's state of health and its evolution [48]; the “Sentinelles” network, which monitors the health of nineteen pathologies (infectious, seasonal, etc.) [49]. In June 2020, the France 3 media highlighted the fact that neither the SPF/FPH nor the Agence Régionale de Santé ( ARS/RHA , Regional Health Agency) publishes epidemiological data on a commune-by-commune basis. Daily reports are limited to deaths occurring in hospitals, as well as in the Etablissements d'Hébergement pour Personnes Âgées Dépendantes (EHPAD/REFDE, Residential Establishments For The Dependent Elderly), for which deaths are aggregated and transmitted only at regional level [50]. At global level, the Organisation Mondiale de la Santé (OMS/WHO, World HealthOrganization) also produces mortality atlases for specific countries and regions. These mortality atlases illustrate and present data on causes of death, mortality rates, as well as temporal trends and the development of these axes in different countries and regions of the world. This enables the OMS/WHO to monitor global mortality, assess progress in public health, as well as develop strategies to reducing these mortality rates [51]. These national and internationalatlases are important tools for decision-makers, be they politicians or specialists involved in health, researchers or professionals at all levels working to improve public health. 2.3. The POLESAT health geomatics project and its latest tool, the uMap mortality atlases POLESAT stands for " Pôles Sanitaires " with the capitalized acronym designating our overall project, an innovative e-geo-platform that will evolve over time, aimed at both the general public and professionals [52, 53]. POLESAT currently comprises four modules: an e-atlas dedicated to the visualization of hospital care supply and demand [54] a medicalized geographical support dedicated to patient orientation and programmed medical choice [55] a prospective health planning tool with variable scenario simulation geometry, available in two versions: "e-PoleSat-démo", a public version, and "e-PoleSat-métier", a private version [52, 53]. The all-cause mortality uMap atlas environment is the fifth POLESAT module. 2.3.1. The new mortality atlas environments of the overall POLESAT project All-cause mortality atlases in the uMap environment are hosted on the umap.openstreetmap.fr server [56, 57, 58] and for the Groupement des Hôpitaux de l'Institut Catholique de Lille (GHICL ,Group of hospitals of the Institut Catholique de Lille) on the ghicl .net server. These environments have been invested over time by the team of developers and geomaticians. Working in two environments, requiring different geomatics and development skills, enabled us to compare and reflect on the technical approaches used and the graphic and cartographic renderings obtained. Comparisons focused on database processing, development and geomatics tasks with remote server management, and geomatics tasks without remote server management [59]. These comparisons concerned two types of "workflows"[7], one based on a "software chain" and the other on "a single software environment": · The first workflow produces two types of data pipelines [22 , 23] Cartographie et Dessin Assistés par Ordinateur ( CAO-DAO/CAMD, Computer Aided Mapping and Drawing) and SIG/GIS including a source code editor; an Environnement de Développement Intégré (IDE ,integrated development environment) for R [60] (a programming language for statistical calculations and graphics), spreadsheets etc. ; · The second workflow is internal to the SIG/GIS QGis software environment, integrating specialized task automation tools [61]. Finally, the literature lists some very positive comparisons and reflections on uMap such as (1) data management in the context of collaborative mapping; (2) data confidentiality in the context of activist mapping; (3) contributor management in the context of "crowd mapping" [16]; and (4) open-source philosophy in the context of teaching collaborative mapping online [59 , 62]. This led us to an overall assessment of the advantages and disadvantages of the two working environments, and to ideas for future versions. uMap is the only environment presented in the article. 2.4. Reorientation of the original idea of an atlas to combat the covid-19 pandemic The atlas of all-cause mortality at regional and French national levels began during the period of the first French containment of the covid 19 health crisis. Our primary objective was to contribute, like so many others and within our means, to the fight against the pandemic. To do this, we wanted to show customers, patients, and hospital professionals the lethal epidemic impact and excess mortality of covid-19 at a fine territorial mesh, in real time. However, after monitoring the situation (both scientifically and in the media), and re-evaluating our resources (both human and technical), we reoriented our primary objective towards: 1- The use of all-cause disease mortality data [63] which offers a number of advantages: the data are finely disseminated at the local level; access is open (open-data); they make it possible to bypass young covid databases with non-negligible biases, such as "errors in coding deaths by covid-19 diagnosis" [64]", 2- A geomatic rather than an epidemiological approach: since we wanted to test geomatic technologies for relevant visualization of statistical mortality indicators and dissemination of geolocalized information. In addition, the scientific and media monitoring carried out as we progressed through the covid-19 crisis in 2020 showed that: 1- French epidemiological data on the incidence of covid-19 became unavailable at the commune level, and the few covid cluster maps available at the commune level eventually disappeared from the French media. Belgium disseminated covid incidence per 100,000 inhabitants at the commune level [65], 2- Many French studies and their associated maps were based on highly aggregated covid-19 open data [66]. [1] The Open-Source world is characterized by the freedom to use, copy, study, modify and redistribute. The challenges of free software are 1-liberty, 2-perpetuity, 3-interoperability, 4-quality-reliability-security. [2] Free geographic information is a formalized product produced within a non-institutional framework, in a specific ecosystem, according to the tried and tested methods of scientific work, namely the rectification process, the reproducibility of results and the verifiability of information. [3] An open standard is any interoperable communication, interconnection or exchange protocol and data format whose technical specifications are publicly available and unrestricted in terms of access and implementation. [4] Here's a definition of a data pipeline: A data pipeline encompasses a series of actions that begin with the ingestion of all raw data from any source, to rapidly transform it into data ready for exploitation. [5] "OGC standards are the cement of geospatial information interoperability, used by thousands of organizations worldwide and represented in millions of lines of code". [6] As Roger Johansson said in 2004: "Web standards are technologies established by the W3C and other standards bodies for creating and interpreting Web content. These technologies are designed to create documents that are durable and accessible to all. [7] "Workflow" is an earlier IT concept, closely related to "pipeline". Workflow is more related to human activity than to machine activity. However, processing chains within software environments are called workflows, as in QGis for example. 3. Material and methods 3.1. Material Our data sources, software (open, free, or private), formats and standards are listed below, along with a brief list of software associated with our early project watch. Open statistical data from the French National Institute for Statistics and Economic Studies (INSEE) Data from the Institut National de la Statistique et des Études Économiques ( INSEE , the French National Institute of Statistics and Economic Studies) are available in spreadsheet format (. csv or .xlsx ). Three gross datasets were extracted in May 2020, covering the period from March 1 to May 15, 2020, inclusive (1st French containment of the Covid-19 crisis): · Deaths by commune representing all-cause mortality of the municipal population (2017 census) for the three years from 2018 to 2020 [63, 67, 68]. · The nomenclature of communes, district (“arrondissements”), departments and regions and the Code Officiel Géographique ( COG/OGC, Official Geographical Code) (2020 vintage) maintained by INSEE [69]. · The 2017 population census data are made up of two files: one in csv format "Communes.csv" listing the population of mainland France and the Départements et Territoires d'Outre Mer (DOM-TOM) renamed in ( Départements et Régions d'Outre-Mer (DROMs/OSDRs, the OverSeas Departments and Regions) excluding Mayotte, and one in Excel format "mayotte-RP2017.xls" for Mayotte only [63, 67, 68]. See the deathDataset repository [63, 70] in [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip] and [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/SF1_supplementary_file_1.zip] Open geolocalized data from the Institut National de l’Information Géographique et Forestière (IGN, National Institute of Geographic and Forest Information) The Institut National de l’Information Géographique et Forestière (IGN, National Institute of Geographic and Forest Information) raw basemaps for the year 2020, are so-called shape files (Esri's proprietary .shp file) which are reworked and adapted for mapping in Philcarto with the help of the “Eclat” utility [71, 72]. There are two tables of communes in France, one of which is lightened by a less precise mechanism for defining communal boundaries: · The "admin-express 2020" database at the local level for the whole of France and by territory [69] updated December 17, 2019, files dated March 25, 2020 ADE, license Etalab 2.0. · The referenced set “ADE updated to May 18 , 2020”, which covers the geographical area of the Départements et Régions d'Outre-Mer (DROMs/OSDRs, the OverSeas Departments and Regions ) (still called DOM_TOM in 2020), with the standard ( EPSG _5490) and the universal file format .mid, .mif. Other open geolocalized data published in “data.gouv” website · Geolocalized data layers for uMap with copyright (©OpenStreetMap contributors, ODbL license) [73-77]. Data processing software (open: free, open-source, public domain) · SQLite version 3.39.4, is a database used to process all data from INSEE [78, 79]. · The R project is a language and software environment dedicated to statistics and graphics [60]. It is multi-platform software, licensed under the licence publique Générale (GNU Non-Unix/ GNU's Not Unix general public license). The R project has a large ecosystem of extensions (packages), an Environnement de Développement Intégré (IDE , integrated development environment) and a version management system (Git and Svn). · The RStudio environment includes the R: v4.2.2 program, the libraries (plyr[8]; openxlsx) and RMarkdown, which provides a report of the processing carried out by R (see Figures 1, 2 in section 3.2.) [80, 81, 82 , 83 , 84]. Cartographie et Dessin Assistés par Ordinateur ( CAO-DAO/CAMD Computer Aided Mapping and Drawing) software (open or free) · QGis SIG/GIS , v3.18.1-Zürich [61] is used for our modeling and workflow processing. It integrates code versions: 202f1bf7e5 and the following libraries: Qt: v5.11.2; GDAL: v3.1.4 [85] GEOS: v3.8.1; CAPI v1.13.3 and Proj: Rel. 6.3.2, the Système Géodésique Mondial (WGS 84, World Geodetic System 1984); as well as Project- CRS -ellipsoid_ EPSG _7030; Système de Référence des Coordonnées (CRS , Coordinate Reference System) and Project- CRS EPSG _4326 May 1st, 2020. · Philcarto v2021.d [86], this mapping software combines a macro language ( .pmc format to automate the opening of work environments) and several export formats (.ai; .emf ), including the main georeferenced .kml format. · ”Eclat” [71, 72] a utility for processing shapefiles in shapefile. shp format or universal files in .mid and .mif formats. · MapSharper (JavaScript) for optimizing data in GeoJSON format [87]. · Geojson .io , an online editor for .geojson files [88]. · XnView and XnConvert for batch image processing, combining a scripting language like Langage de Requête Structurée (SQL , structured query language) [89, 90]. · IrfanView for image cropping with batch processing functions [91]. Private image processing or Dessin Assisté par Ordinateur (DAO/CAD , Computer Aided Drawing software) · Photoshop© v7.0 (including .atn scripts) for automated legends in uMap [92]. · MapInfo© SIG/GIS v7.0 - old version is used to manage and retrieve basemaps. · Paint© [93] and Paint .Net combine the Visual Basic pour Application (VBA, Visual Basic for Application) language to automate drawing tasks [94, 95]. · SpreadsheetLight© function library used by the "Eclat" utility [96]. Free and open-source collaborative platform · The uMap interoperable software package (v1.2.1) [97] is a webmapping environment used to represent our France-wide atlases by variable. uMap allows the creation of maps with OpenStreetMap layers . It is a collaborative mapping software platform licensed under the Licence Publique Faites en ce que Vous Voulez (LPFVV/ WTFPL, Do What The Fuck You Want To Public License) [98] (free software: redistributable and modifiable; Gratuit/Libre et Logiciel de Source Ouverte (GLLSO/FLOSS, Free/Libre and Open-Source Software) [59, 99, 100 ]. uMap is hosted on the French server openstreetmap.fr [59] with the domain name umap.openstreetmap.fr [57, 101]. It is a Django project maintained by a community of developers. Globe · Google Earth Pro© is used to view and export the kml format [102]. Code management software (free, GitHub repository) · Visual Studio Code, under Microsoft-MIT-license [96]. · View map extension for VSCode, published by Random Fractals Inc [103]. · Git Bash for Windows (script command-line interpreter), provides a Bourne-Again SHell (BASH) emulation for running Git (a distributed version control system) from the command line [104]. Consortium Géospatial Ouvert (OGC, open geospatial consortium) standards · Keyhole Markup Language (KML) , Google's proprietary format, now a Consortium Géospatial Ouvert (OGC open geospatial consortium ) standard. The KML specification is associated with only one projection, EPSG _4326 [105-110]. · GPX , is a Global Positioning System ( GPS) eXchange Format, to export format from uMap environment and OGC standard [7, 26]. Internationale Association des Producteurs de Pétrole et de Gaz ( PPG/OGP , International Oil and Gas Producers Association standards · The EPSG _4326 projection is associated with the KML standard [31]. Proprietary, open, standardized and brand-specific formats · Shapefile (Esri©) [111, 112]. proprietary, non-standardized, brand-specific formats · GeoJSON (RFC 7946 - OpenStreetMap): see the guides and editors of this format [88, 108, 109, 113]. · model3 (modeler® QGis ) [114]. · Atn (script-modeling, macro recording from Adobe Photoshop© PS7) [92, 115]. · Pcm (Script-modeling, Philcarto macro recording) [86]. Computer-aided drawing (DAO/CAD) software watch (free, open), collaborative platforms - high learning curve A monitoring period is essential for any project and has led us to partially test the following software [116]: · GIMP-V2.1 combines multilingual support (including Python and Scheme) with image manipulation. It is an alternative to proprietary Adobe Illustrator [117]. · Inkscape combines a Simple Inkscape Scripting extension and the Python language to automate repetitive drawing tasks [118, 119]. · Mapbox© has been tested to geolocalize and convert .svg to .geojson data before import into uMap [120]. 3.2. Methods We begin by presenting the software architecture describing the data processing stages (see Figure 1: Synthetic (a) and detailed (b) webmapping architectures of uMap atlases for France), including (Data transformation diagram of Figure 2). In our study, data management (ingénierie des données) is an essential process carried out in different software environments, involving data extraction, cleaning and ordering [121], involving specialized concepts and processes such as data pipelines, databases and the Extraire, Transformer et Charger ( ETL, Extract, Transform, Load) process. Our methodological approaches are schematized to enable better monitoring of data pipelines “step 1- without webmapping” or “step 2 - with webmapping”. “For step 1 - without webmapping”, can be seen the data pipeline diagram for transforming . kml into a . geojson file in Figure 3 and schematics of processing automation in QGis in Figures 4 and 5. “For step 2 - with webmapping”, can be seen the resizing processing schematic in Mapshaper in Figure 6; the double caption processing diagram in CAO-DAO/CAMD and uMap in Figure 7; the thumbnail processing diagram in CAO-DAO/CAMD software in Figure 8. Finally, the Photoshop processing automation diagram is shown in Figure 9. 3.2.1. Atlas web architecture 3.2.2. INSEE death data engineering All data sources (see section 3.1. Material) share a common identifier: the INSEE commune code, consisting of five alphanumeric characters (e.g. 2A014) . Structured Query Language (SQL ) is used to analyze and process data in SQLite [78, 79]. All processing is automated, with the exception of one that remains in manual mode. The simplified data transformation diagram is shown in Figure 2 below. Data in the form of . csv files are imported into SQLite and then combined to produce an "insee_dc" table. An analysis of the fields (or variables) shows that days and months of birth are not systematically filled in. We have calculated the age of each person, with the exception of those for whom the day or month of birth was not provided. The data is then aggregated by INSEE communal code, calculating the average age (excluding date with birth not provided), the 2020 mortality, the average mortality over the two years 2018 and 2019, and the mortality difference. The final step in the Figure 2 process is to associate the death data with the reference 2017 municipal population. The database thus created " insee_dc_pop " is exported in . csv format. Table 1 below shows the reference structure of the data corresponding to the mortality variables. Table 1 Structure of death database (source INSEE ) Field Description Example Id_com Municipal code “01004” Id_dep Departmental code “01” Id_reg Regional code “84” Label_com Municipal name Ambérieu-en-Bugey Muni_pop_2017 Municipal population no. 2017 14,035 Dc_2018 No. of deaths 2018 36 Dc_2019 No. of deaths 2019 42 Dc_2018_2019 Average no. of deaths 2018-2019 39 Dc_2020 No. of deaths 2020 49 Mortality_diff. Difference of deaths (dc_2020 minus dc_2018_2019) 10 Average_age Average age at death 82 SMM_10000 Excess mortality/undermortality per 10,000 inhabitants 7 3.2.3. INSEE data engineering in R, Excel for Philcarto Several stages of data pre-processing and statistical calculations in R, Excel and Philcarto: In the pre-processing stage (see in 3.2.3., Table 2 -section lines [1] to [21]), the output file " insee_dc_pop.csv" from Figure 2 requires further pre-processing steps in R and Excel, essentially for the addition of statistical variables, and for the addition and renaming of fields; it can then be used in Philcarto and in the processing chain to instantiate atlases in uMap. Compared with (Table 1), the output file includes modified, renamed, created, and reordered variables. Suffixes added to variable names enable them to be typed directly by Philcarto [72, 86]. In our database (Table 2), the suffix _R_ indicates a report variable and _N_ indicates a nominal variable. After processing in R, the output file in .xlsx format now contains 24 variables. In the statistical calculation stage ( see in 3.2.3., Table 2, line sections a [25] and a [29]), Excel spreadsheets are supplemented by the additional statistical variables Chi-square and Chi-square with logarithmic transformation "Khi2LnFr", the methods of which are described in (section. 3.3.2: Calculation of the four atlas variables). Variables are sometimes renamed (see in 3.2.3., Table 2, line sections b [25] and b [29]) to bring them into line with the Philcarto file header standard: for example, the variable " (Obs - Theo) ^2/Theo_Fr" or Khi-deux brut is renamed “ DENSI”. At the " FeuTricolore "(TrafficLight) classification stage (see in 3.2.3., Table 2, lines [31] and [36]), a new series of Excel processes is used to classify the values of the " Khi2LnFr " variable to give the fourth " FeuTricolore " (TrafficLight) atlas variable associated with three colors, and to create the sub-national maps in Philcarto using the " Surmortalité_feu_tricolore_umap " (Excess Mortality_TrafficLight_umap) variable. This is also the stage at which the input file insee.xlsx is enriched with information variables such as " Surmortalite_feu_tricolore_umap " (Excess Mortality_TrafficLight_umap), recognized directly by the suffix "umap", which is used to interactively display the communal tooltips (on hover) of uMap atlas instantiations. At the Philcarto processing stage (see in 3.2.3., Table 2, line sections [31] and [36]), we present the following files as input to the software: spreadsheet-insee.xlsx (France as a whole, ~36,000 communes online) and BaseMap .shp (georeferenced shapefile for the region or France as a whole) [72, 86]. As an output, we obtain an initial cartography associated with our four atlas variables designated "SMM_10000", "DENSI", "Khi2LnFr" and " FeuTricolore " (TrafficLight). The " FeuTricolore " (TrafficLight) atlas variable is named " Surmortalité_feu_tricolore " (Excess Mortality_TrafficLight). The output formats used are: .ai; .pmc; .emf; and .kml. The .kml format is georeferenced with the Système Géodésique Mondial (WGS 84, World Geodetic System 1984 ) – EPSG _4326 [29, 31, 105, 106, 107, 108, 109, 110] . The Keyhole Markup Language KML or kml format enables interoperability and the start-up of the processing chain in other software environments, with the aim of obtaining a ready-to-use map in . geojson format for the uMapOSM.fr remote server (see Figure 1 and Figure 3 ) . The maps are called choropleth or color range maps and are systematically accompanied by a legend discretized using Jenks' method (see section 3.3.3) [122, 123]. Table 2 INSEE variables transformed after processing in R and Excel Group Line numbers Variable names per line 1 [1] "Iden" "Iden_1" "Id_dep" "Id_commune" "Label_commune" [6] "Population" "Muni_population" "Dc_2018" "Dc_2019" "Dc_2018_2019" [11] "Dc_2020" "Excess mortality" "Average_age" "SM_10000" "SMM_10000_R_" [16] "status" "Period" "Id_2020_reg" "Id_2020_dep" "Label_2020_dep" [21] "Label_2020_reg" "Period_N_" "statut_N_" "Excess_mortality_R_" 2 a [25] "Dc_2020_Theo_Fr" "(Obs - Theo) ^2/Theo_Fr" "Dc_2020_Theo_reg" "(Obs-Theo)^2/Theo_reg" a [29] "Khi2LnFr" 3 b [25] "Dc_2020_Theo_Fr" "(Obs-Theo) ^2/Theo_Fr" "DENSI" "Khi2LnFr" b [29] "Dc_2020_Theo_reg" "(Obs-Theo) ^2/Theo_reg" 4 [31] "Id_region" "Khi2LnFr" "10_classes" "Class" "TrafficLight" [36] "Class_1" "Excess_mortality_TrafficLight" "Excess_mortality_TrafficLight_umap" Legend for Table 2: Four groups (g) and line sections (l): G1, lines 1 to 21: 24 variables obtained from R. G2, lines a25 to a29: variables N° 25 to 29 obtained in Excel. G3, lines b25 to b29: variables N° 25 to 30 prepared for Philcarto. G4, lines 31 to 36: variables N°31 to 38 prepared for the “ FeuTricolore” (TrafficLight) map, workflows, and tooltips. Suffixes for G1 variables are shown in bold. Note that we have written the Chi2 variable, "Khi2 and Khi2LnFr" in our databases. 3.2.4. IGN base map data engineering Using files from the Institut National de l’Information Géographique et Forestière (IGN, National Institute of Geographic and Forest Information) (see Material section 3.1.), we created base maps at the scales of the DROMs/OSDRs , large regions (grouped regions) and the arrondissements of Lyon, Paris and Marseille, and georeferenced them the Système Géodésique Mondial (WGS 84, World Geodetic System 1984) for GPS [29, 31]), resulting in a .kml map output for CAO-DAO/CAMD software (see material section, Figure 1: Synthetic (a) and detailed (b) architectures of uMap atlases for France and Figure 3: Data pipeline, geomatic processing software system - step 1 - without webmapping). For reasons of vector file size, we have broken down the .shp shapefile of France's communes into 4 major sub-national commune regions using the "Éclats" tool [71, 72]. One or more thicker administrative boundary layers are added to the base map. We produce as many . geojson layers for uMap as there are choropleth maps produced for an atlas. That is 12 geojson layers for the “ FeuTricolore” (TrafficLight) atlas (4 large regions, 3 arrondissements and 5 OverSeas Departments and Regions DROMs/OSDRs . For further information on map background engineering, please refer to the author's works [71, 72, 124-126]. 3.2.5. Geomatic data engineering For the sake of clarity, we present our data pipelines and workflows mainly in schematic form. Geojson creation step for uMap (step 1 without webmapping) The .kml (direct) file produced by Philcarto is our first file to undergo geomatic processing flows (workflow) via a data pipeline (see uMap environment sections 3.4.1: Step 1- geomatics without webmapping and 3.4.2: Step 2- geomatics with webmapping of uMap environments). In several software packages, transformations made to the “direct .kml ” result in the final output file in .geojson format ( see 3.4.1.1., Figure 3: Data pipeline, geomatic processing software system – step 1 without webmapping). The workflow in QGis is based on two input files “GE_vscode .geojson” and the “spreadsheet .csv” (utf8) , and on the workflow model generated using the "modeler" tool, which generates files in . model3 format . These models enable batch processing. From QGis, we obtain the output files, in the form of maps and enriched data in . geojson format, ready for uMap. See, both figures: (Figure 4: Workflow QGis (.model3 ) part 1-obtaining the TrafficLight. geoson map for uMap – step 1 - without webmapping; Figure 5: Workflow QGis ( .model3 ) part 2 - batch processing – TrafficLight. geoson for uMap- step 1- without webmapping) For example, the " Surmortalite_feu_tricolore_umap " (Excess Mortality_TrafficLight_umap) variable is automatically transformed to display structured tooltips when the mouse is moved over the " FeuTricolore " (TrafficLigtht) atlas. Legends creation stage (step 2 with webmapping) The .geojson map obtained from processing the .kml format (in step 1 without webmapping) is uploaded as an input file to the uMapOSMfr remote server . Once in place, the .geojson map is managed using the uMap "layer window" manager and "dashboard" (see in section 4. Results, Figure 11: uMap atlas by TrafficLight, France, address URL the atlas. ) The legend associated with each .geojson map is displayed using the legend creation and formatting data pipeline shown in Figure 7. The workflow for creating caption thumbnails is shown in Figure 8, while the workflow for formatting captions (automated in Photoshop using “script. atn ”) is shown in Figure 9. .png or .jpg legend files are uploaded to a remote server at the University of Lille, as these formats cannot be hosted on the uMap OSM server, which is reserved exclusively for .geojson files [22 , 23]. (Figure 12: uMap atlas by TrafficLight- zoom on the south-west regions - Bayonne town tooltip. URL address of the atlas. ) 3.3. Step 1 method - geomatics without webmapping 3.3.1. Choice of scales and meshes For our mortality atlases, we have chosen to present a France-wide analysis included the OverSeas Departments and Regions DROMs/OSDRs at a fine territorial grid (commune or arrondissement of the cities of Lyon, Marseille and Paris) and also to complement existing studies that mainly process information at coarser grids [1]. 3.3.2. Calculation of the four atlas variables Statistical calculations are made using death data from the French National Institute of Statistics and Economic Studies ( INSEE ) to create atlases (see section 3.2. Methods) based on the following four variables: The " SMM_10000 " excess- and under mortality variable Formula (a) : This variable is calculated from data on communal deaths from all causes, with three reference years: 2018, 2019 and 2020. The numerator (the difference between deaths in 2020 and the average number of deaths in 2018 and 2019) is divided by the municipal population (2017 and per commune); this ratio is then multiplied by 10,000. This variable highlights over- or under-mortalities per 10,000 inhabitants. The " DENSI " variable for a Raw Chi square (Raw Chi2 ) Formula (b): We calculate a density of deaths, which is the ratio of the number of national deaths to the total French municipal population (metropolitan or excluding Départements et Régions d'Outre-Mer ( DROMs/OSDRs, the OverSeas Departments and Regions) ) (~64M); this ratio is then brought down to the municipal population, enabling a comparison to be made between actual deaths and theoretical deaths. In this way, we can highlight over- or under-mortalities, which are more relevant and significant in terms of probability compared with the national average. The "Chi2LnFr" variable for a Chi square Ln Fr. Formula (d) : Using a logarithmic scale, the “ DENSI ” variable is transformed with new values that informs directly about levels of excess- or under mortality compared with the national average. The "TrafficLight " variable The values of the Chi2LnFr variable or even noted (Khi2LnFr) are grouped into three color classes (green, orange, red) using Jenks' discretization method in Philcarto. The semantic interest of the mapped “ FeuTricolore” (TrafficLight) variable provides humans with statistical information that is simplified (three classes) and directly understandable (given the strong symbolism of a traffic light). 3.3.3. Jenks map legend classes Every map legend is based on a statistical method for discretizing the data. We have chosen Jenks' method, also known as the "natural threshold method", which takes as input a defined number of classes. This method maximizes the variance between classes and minimizes the variance within classes, thus making each group of geographical units (communes or arrondissements) more homogeneous (compared to the other methods proposed by Philcarto). Computer calculations can be time-consuming, but the result is optimal. The produced maps are fairly objective, if the number of classes is greater than 5 . This method is better suited to multimodal variables possessing a large enough number of individuals to be able to bring out thresholds. It waś developed by G. F. Jenks, an American geographer and cartographer, in the 1960s [127, 128]. 3.3.4. Terminal repository - ten national classes, sub-national class bases A map of the whole of France is created for each variable (three maps), using Jenks' method and a legend of 10 classes. The boundary values of the 10 classes are stored in a repository and used to create thematic maps (of the same variable) on a sub-national scale (region, large region, district (“arrondissement”) and the OverSeas Departments and Regions (DROMs/OSDRs ) . E xample of ten-class frame of reference See Table 3 below, showing the boundaries of the 10 classes for the variables " FeuTricolore " (TrafficLight), and "Khi2LnFr". For the " FeuTricolore " (TrafficLight) variable, we essentially target excess mortality compared with the national average. Thus, 3 color classes green, orange, and red represent increasing excess mortality: (Green => 0 - 1.83, means close to the average to 2 times above the average; Orange => 1.83 - 3.74, means 2 to 4 times higher than the average and Red => 3.74 - 11.07, means 4 to 11,07 times higher than the average). Table 3 : Ten-class frame of reference based on Jenks discretization (variable: Khi2LnFr) Class Khi2LnFr interval No. of communes 4 TrafficLight 10 [+3.736, +11.075] 581 Red 1 9 [+1.826, +3.736[ 1724 Orange 2 8 [+0.747, +1.826[ 4629 Green 3 7 [-0.074, +0.747[ 7569 Green 3 6 [-0.844, -0.074[ 8409 5 [-1.712, -0.844[ 7070 4 [-3.056, -1.712[ 3466 3 [-5.215, -3.056[ 976 2 [-8.78, -5.215[ 340 1 [-21.245, -8.78[ 69 Legend for table 3 Four out of ten classes belong to the “ FeuTricolore” (TrafficLight): 1 Red [3.74, 11.07], 4 to 11 times higher than average. 2 Orange [1.83 - 3.74[, 2 to 4 times higher than average. 3 Green [0, 1.83[, close to average to 2 times above average. 4 "communes" or number of towns identified (France as a whole). 3.4. uMap environment and atlases by variable - two-step geomatics action plan In order to obtain an atlas environment (for each variable) instantiated in uMap, we followed a two-step geomatic action plan: "step 1, geomatic processing without webmapping" based on the construction of maps in . geojson format outside uMap , then "step 2, geomatic processing with webmapping" based on the creation of legends with the help of uMap. 3.4.1. "Step 1- geomatics without webmapping See in 3.2.1., Figure 1: Synthetic (a) and detailed (b) webmapping architectures of uMap atlases for France 3.4.1.1. Step 1 data pipeline description Various software and processing methods are used to obtain instances of the “ FeuTricolore ” (TrafficLight) atlas in uMap. In order to obtain files in .kml format and then in .geojson format ready for uMap, we processed the information in a software suite producing a data pipeline [22]. To illustrate the complexity of the data pipeline, we present its schematic in Figure 3 below. It shows a succession of tasks ordered in various software packages which are mainly of the following types: Cartographie et Dessin Assistés par Ordinateur (CAO-DAO/CAMD Computer Aided Mapping and Drawing), SIG/GIS , Globe, source code editors, spreadsheet, etc. (see Material section 3.1.). Data pipeline diagram step 1 - from .kml to .geojson for uMap Following the progression in Figure 3, we can see that maps in .kml file format are transformed to obtain the final .geojson format . The . kml map thus undergoes a series of processes in several software environments (Google Earth, VScode and QGis, the .geojson file editor) until the final map.geojson file is obtained at the end of "step 1: geomatics without webmapping". For further information, please refer to Figure 1. Sometimes the pipeline is partially automated, otherwise it is fully automated, as in the two QGis workflows Figure 4 and Figure 5 below [61]. The task processing chain is carried out within the "same software environment". For this example, the "graphical modeler" is used [114] to model the automation of an entire geomatic processing chain, starting with an "input", before obtaining the output .geojson file (for uMap). For the creation of a map of France, two files were created from the graphic modeler ( .model3 ): "file: processing chain" and "file: batch processing" (see in section 3.2.1., Figure 1). QGIS workflow diagrams - processing direct .kml to obtain geojson file for uMap - step 1 without webmapping In QGis, the workflow is broken down into two parts in Figure 4 and Figure 5 below. 3.4.2. "Step 2- geomatics with webmapping of uMap environments Successively following step 1, the final map .geojson file obtained is ready for use and becomes the "input file" in step 2 geomatics with uMap webmapping [99]. Please note: The .geojson file obtained in step 1 may not be imported as is, due to size constraints imposed by the OSM server administrator. In this case, for certain files, we proceed to lighten them by using the MapShaper tool, enabling us to reduce the size of the geojson file [87, 129] by reducing the definition of vector shapes. Step 2 data pipeline diagram with webmapping - .geojson lite for import into uMap Figure 6 legend: A distinct shape is associated with each of the two types of information: diamond: processing; rectangle: input-output-result. 3.4.2.2. uMap platform features Once the .geojson file has been imported onto uMap's remote OpenStreetMap ( OSM ) server, to address the complexity of processing in step 2, we have again diagrammed the data pipeline shown in section 3.4.2.3., Figure 7: Double caption data pipeline…)” below; see also in section 3.2.1., Figure 1: Synthetic (a) and detailed (b) architectures of uMap atlases for France”. In the uMap environment There are few or no lines of code to control the uMap dashboard environment. The map display is set in the Interface Utilisateur (IU/UI, the User Interface) or dashboard. Once the Localisateur de Ressources Uniformes (URL , Uniform Resource Locator) address has been launched in the browser, uMap loads each file of the major regions, arrondissements and DROMs/OSDRs in turn. The final display of the " FeuTricolore " (TrafficLight) map can be viewed at the scale of the whole of France. Creation of a map settings template: based on the first settings made in the dashboard for a map in the “FeuTricolore” (TrafficLight) atlas (i.e., the parameters for displaying the map, its titles, font and legend), we generate a settings template. This template is used to automate and save time when integrating and displaying other variable atlases and their maps in .geojson format in uMap. Open data contour and health information layers: provide administrative boundaries and health information. They are uploaded in .geojson format to the uMap OSM .fr server. They are managed using the uMap layer manager (see Material section 3.1.). To host the legends created and associated with each .geojson map of the major regions, arrondissements and DROMs/OSDRs of the France atlas, the creation of ownership of a hosting server is required. Legends in .jpeg or .png format cannot be uploaded to the same remote server as the atlas .geojson files. The uMap atlas is secure from the moment it is created, simply set to secure uMap - security maintenance of the OSM remote server is managed by four administrators from the OSM association, with three servers currently hosted by OVHcloud France. Map data in .geojson files are not encrypted on the server. The servers are non-compliant with the Règlement Général sur la Protection des Données (RGPD/GDPR , general data protection regulation) [130]. We are not hampered by the few weaknesses of the uMap environment described in the literature by [59, 62] notably that the platform is not very collaborative, which has no impact on our work. 3.4.2.3. Step 2 data pipeline description To address the complexity of geomatic processing for atlas embellishment in uMap, we have schematized the data pipeline that creates and associates a double legend to each .geojson map. Double caption data pipeline diagram - step 2 with webmapping Figure 7 includes three blocks of procedures numbered from 1 to 8. Following the progression, we start with a file from Philcarto, this time in .emf format, which at the end of the chain results in a double map legend with QRcode in .png format; these legends are all hosted on a remote server on the univ-lille.fr domain and are associated with each .geojson map on the uMap OSM server. Thumbnail workflow diagram - preparing legends for step 2 uMap webmapping Workflow diagram in Photoshop - double legend with and without QRcode -step 2 with webmapping [8]The plyr library is replaced by "dplyr (for data blocks)" and "purr (for lists)". 4. Results Here, we mainly present the results of the Philcarto cartographic environment (step 1) and the results of step 2 geomatics with webmapping (see section 3.4.2: Step 2- geomatics with webmapping of uMap environments). In step 2, each map of the regions making up the France atlas by variable is imported into uMap's remote OSM server. The visual enrichment of the atlas, by displaying legends and tooltips, also requires two processing chains: the first internal, using the uMap dashboard, and the second external, using various environments such as Xn View-XnConvert and Photoshop (see material section). In this article, we mainly present the results of the “ FeuTricolore” (TrafficLight) atlas and provide the three uMap links to access the online atlases. 4.1. Philcarto geomatics environment (excluding webmapping) - .kml maps – “FeuTricolore” (TrafficLight) As a first part of the results, we show below, mainly for step 1, the cartographic results of the Philcarto environment, enabling us to obtain the first exported map in . kml format (see in section 3.2.1., Fig. 1 ). This Philcarto environment consists of a two-part window: the first associates the map result and its legend, and the second, the tools and parameters for map creation. Then, as the visual aspect of the map remains identical, there is no point in showing it again in the intermediate processing environments, until the final . geojson format for uMap (reduced in file size) is obtained. In step 1, our initial map is processed using: VSCode, Google Earth, QGis, geojson.io and MapSharper, with only the spreadsheet data enriched at the end. 4.1.1. Interpreting the FeuTricolore (TrafficLight) choropleth map (In Fig. 10 : Excess mortality map by traffic light (map of atlas) output. kml format – realized with Philcarto), we're looking at a choropleth map and its legend for the " FeuTricolore " (TrafficLight) variable, dressed up with a departmental outline and presented on a French scale. The map covers the period of the 1st covid-19 crisis confinement, from March 1 to May 15 inclusive in 2020. Based on the Khi2LnFr variable, our statistical indicators are transformed into classes in the legend entitled " surmortalité_feu_tricolore " (excess mortality_TrafficLight). There are three classic color classes, for the semantic representation of traffic light, and a pale color (class excluding traffic light) grouping together geographical units close to the average mortality (metropolitan France) or expressing undermortality compared with the average. The number of geographical units included in the map is 34833 (6 communes are not included because their population is zero). 20330 communes (around 58%) are classified as not having traffic lights. The reference French municipal population is 63,639,133. Of the 1,4503 communes classified according to traffic lights (around 42%), 1,2198 communes (around 35%) are in the green class and are positioned in the interval "close to average to 2 times above average" ; 1,724 towns (around 5%) are classified as orange, in the range "2 to 4 times above average", and 581 towns (1.66% or around 2%) are classified as red, in the range "4 to 11 times above average" , which represents an extremely significant excess mortality rate. Below the classes in the " surmortalité_feu_tricolore " “excess mortality_TrafficLight” legend is a bar chart, with the heights of the rectangles proportional to the number of spatial units (34833) included in each modality of the distributed variable, reading from left to right: 581 (2%); 1724 (5%); 12198 (35%) and 20330 (58%). NB The aim of this map is to alert the observer at a glance (visually), and to simplify interpretation for the general public, using semantics known to all. 4.2. uMap geomatics environment (with webmapping) - instantiation of atlas-“FeuTricolore” (TrafficLight) - OSM remote server The geomatic results obtained in "step 1" are used and integrated into the processing chain processes of step 2. Thus, in the second part of the results, we present below, mainly for step 2 uMap, the showcase of uMap functionalities, the results of the display of the " FeuTricolore " (TrafficLight) atlas for the whole of France and a zoom on the south-west region with a tooltip (parameterized in the uMap dashboard) pointing to the commune of Bayonne. 4.2.1. Feature showcase: ergonomics and usability in uMap Our uMap showcase includes our four atlas instances as main outputs. Three URL s provide access to our atlases (one URL per variable) for the whole of France, including Départements et Régions d'Outre-Mer (DROMs/OSDRs , the OverSeas Departments and Regions). There are as many double legends as there are . geojson layers imported per atlas. Atlas access and display times are relatively short. However, loading into uMap is somewhat slower, due to the display of the complete atlas, which loads . geojson layers at commune level for the major regions, arrondissements and DROMs/OSDRs representing around 36,000 communes. Once the atlas is loaded, navigation in the browser is smooth and without slowdown. The dashboard allows choice and configuration of the desired display functionalities and make them available to users from the atlas' uMap interface. If desired, users can contact the atlas designer to obtain the complete data in . geojson format or a copy of the atlas. From the user interface, several functions are accessible, accompanied by icons and/or menu names: zoom + and zoom -, search for a place name, full-screen view, center the map on location, measure distances, change the background map, map editing link to OpenStreetMap etc., see guide to uMap basic functions [ 58 , 62 , 97 , 130 – 133 ]. Printing atlases requires (1)- access to the data sharing functionality and (2)- retrieval of exported map data in . gpx , and . kml formats (for other SIG/GIS or web services such as My OSM atic [ 134 ]) for printing purposes). Finally, uMap is an interoperable platform that enables the export and printing of maps based on Interface de Programmation d’Application ( API , an Application Programming Interface) [ 59 ]. 4.2.2. Instantiating the “FeuTricolore” (TrafficLight) atlas - uMap world map via OSM The uMap environment in Figs. 11 , 12 consists of a three-part window The first part combines the visual aspect of the France atlas with several tool-buttons on the left of the screen, including our main one for managing and displaying layers (or uploaded regional . geojson maps), as well as searching for data via the "Browse data" link. The second part of the screen, on the right, is the legend tab, where the legends for each layer in the atlas are integrated. The third part represents the window banners displaying the map title, designer and properties [ 131 ]. Here, for the period of the first French confinement, we see the " FeuTricolore " (TrafficLight) atlas, dressed with departmental and regional outlines and legends associated with regional . geojson layers on a remote server. In this way, the " FeuTricolore " (TrafficLight) atlas not only displays excess mortality information thanks to the colors displayed on the map and map legend, but also shows structured information in tooltips set from the dashboard. Here are the URL s for viewing the atlases of the " SMM_10000 " variable: atlas URL and the " DENSI " variable: atlas URL. 5. Discussion The division of tasks between authors gave us the opportunity to invest in various geomatics and IT development environments for the creation and visualization of our atlases. We will discuss some of the main points of the two-step geomatics action plan without and with webmapping, such as the data engineering carried out in step 1 on the collection, cleaning and ordering of death data, and such as the processing flows and optimization for the creation of legends in step 2. 5.1. Discussion - INSEE death data engineering Three selections were made from the death data sets of theFrench National Institute of Statistics and Economic Studies ( INSEE ). 1-Our first reference is the 2017 census population, which aggregates data from 2015 to 2019. Explanation: communal censuses, managed by INSEE , take place every year on a fifth of communes with fewer than 10,000 inhabitants and every year by a survey on a sample of 8% of the population for communes with more than 10,000 inhabitants [135]. The population for a reference year N aggregates the data for the five years N-2, N-1, N, N+1 and N+2. Finally, INSEE indicates that the error induced by the survey method is less than 0.02%. The reference population (2017) chosen for the study aggregates data from 2015 to 2019. 2- The municipal population was chosen for the study. Explanation: two types of population census are determined: the municipal population, which essentially corresponds to usual places of residence, and the population "counted separately" , which also includes other types of residence (second homes, students, etc.) [136] so to be able to compare population data from one year to the next without double counting: i.e. without counting duplicates in the population. 3- We have approximated the average age of death in the face of missing values. Explanation: some days or months of birth are missing (0.7%). However, this average age could be refined by defining a month and a day of birth (6 th month of the year for the month and 15 th day of the month for the day). Our approximation is considered acceptable for the present study, given the low percentage of incomplete data. 5.2. Discussion of the choice of BD data management methods in the RStudio environment In theRStudio Environnement de Développement Intégré (IDE ,integrated development environment), Rmarkdown is an integrated module interwoven with source code (Iterate Programming), allowing users to build the logic followed for at least part of the processing chains in step 1. An example of a “summary program” that processes the input file insee_data_pop.csv (see in sections 3.2.1. and 3.2.2., Figures 1, 2 and Table 1) and outputs a ready-to-use file in .xlsx and .txt extensionsfor mapping in the Philcarto Computer Aided Mapping and Drawing CAO-DAO/CAMD software, and for obtaining a .kml file, can be found in the article's SF1_supplementary_file_1.zip [70] . The recent package for R "Cartography" renamed "mapsf" [122, 137] (not used in the present study), makes it possible to attempt in the future reproducibility of the entire step 1, i.e. this time from the insee.fr input file to the final output file in .geojson format, which represents a choropleth map of the atlas ready for import into uMap. If step 1 can be reproduced in the RStudio environment only, to obtain maps in .geojson format, then the benefits will be manifold, with the help of the RMarkdown report (which enables R programs to be restarted and initial results to be obtained with a single click; or to more quickly/easier correct any statistical errors in the program). 5.3. Discussion of the choice of the uMap webmapping environment We chose an open-source software package that is powerful, easy to use and compatible with other tools, thanks to the interoperability of the software's input and output file formats. For example, uMap supports the download of data in .umap, .geojson (OpenStreetMap proprietary format) and . kml (Google proprietary format standardized by the Consortium Géospatial Ouvert ( OGC, Open GeospatialConsortium) in 2008) formats [1, 59, 99, 105, 131]. It is an ethical environment that does not reuse our own .geojson data loaded on uMap if the sharing rights are not active [62]. uMap is based on Django [138, 139]. The uMap application is fairly simple and intuitive to use, with a dashboard for managing and displaying layers, and numerous integrated functions and resources (world map-type maps) available [130, 132, 133, 140, 141]. We have not encountered any bottlenecks in using the platform. Extensive documentation is available online [131, 132, 138]. The software package is maintained by a dynamic community of developers and geomaticians[9] [62, 97, 99, 100, 142, 143]. uMap is used in many areas of collaborative cartography. The current uMap environment and its remote server managed by OpenStreetMap.fr ( OSM ) hosts maps from all over the world. Using keywords such as "health, mortality", we can find thematic maps if they are shared with the community (sharing enabled). We have not found, if they exist, choropleth maps by filtering by keyword such as "health" and "mortality" [144]. Sharing our mortality atlases on the first French containment in 2020, could be attractive both in terms of the theme and the semantic interest of the “ FeuTricolore” (TrafficLight) atlas to the instantaneous message of excess mortality. Publishing our atlases[10] in uMap, with authorization to share and export data, will be a good test to evaluate the potential of shared information and feedback from the uMap community and platform. It is worth noticing that uMap offers functionalities not yet used for our atlases (1) The " Slideshow" see example in [133], (2) The " Integrate and share a uMap" icon is a feature that allows the map to be integrated within a website, and offers eight export options, including the open standard GPS exchange format (Global Positioning eXchange GPX ) standardized by the OGC for the exchange of GPS coordinates [31, 145]. This sharing function provides a short web address for the map's Localisateur de Ressources Uniformes (URL , Uniform Resource Locator) resources. (3) Finally, the uMap environment can be installed locally, which is ideal for a server dedicated exclusively tohealthcare.Additional IT developments and new functionalities can be added to enrich the local uMap environment. “uMap” provides an Interface de Programmation d’Application ( API, an Application Programming Interface) based on Python and JavaScript languages [138]. (4) Dynamic updating of .geojson files is possible [146]. 5.4. Discussion of two-step geomatic methods for the uMap environment The vast majority of maps produced in the humanities and social sciences are the result of sometimes highly complex processes, involving numerous processing software (or systems) and file formats with limited possible automation of tasks. This is the case in our work, where through our task processing stages, we have been confronted with a pipeline of complex, non-interconnected data that weighs down processing and final execution times for obtaining atlases in uMap. However, even if there is no interconnection between systems [147], software programs produce file formats that enable interoperability when moving from one software environment to another. Below we present a critical analysis of our two-step geomatics action plan with and without webmapping. 5.4.1. Discussion of two-step geomatics action plan with and without webmapping In view of the results obtained, our action plan offers clear advantages, since we use a large proportion of licensed open-source software, free software, licensed open data, as well as environments and functionalities adapted to our needs, that have enabled us to fully automate certain tasks. However, despite these advantages, our two-step geomatic action plan is rather cumbersome, as it relies on a substantial number of geomatic processes. We list below at least four of the main drawbacks encountered: (1) lack of access to the latest versions of proprietary software makes task automation sub-optimal. This is the case, for example, with our (old) version of Adobe Photoshop 7.0 [92]; (2) due to a lack of geomatic functionality, the data pipeline is not automatically interconnected: an example is shown in the following article [147] which looked at a company's system that relied on a PostGIS database, Geoserver and OpenLayers visualizations on a PHP server; (3) the function for managing and creating double legends, to be associated with .geojson layers (maps without legends), is not currently implemented to choropleth thematic maps in uMap: this function is expected in the future. (4) a multitude of . geojson maps has to be produced to instantiate the atlases (by variable) in uMap. 5.4.2. Discussion of design in uMap In terms of design benefits (1) uMap is licensed under Licence Publique Faites en ce que Vous Voulez (LPFVV/ WTFPL, Do What The Fuck You Want To Public License) [98] which characterizes it as free of redistribution and modification, (2) As it stands, this software package delivers the expected results without having to worry about security maintenance of the remote server hosting the .geojson files. The latter is managed by OpenStreetMap or OSM.fr and is an important point for the instantiation of our client-side atlases. (3) We have facilitated the semantic interpretation of mortality by creating the " FeuTricolore " (TrafficLight) atlas, which is considered to be the best messenger to alert us to the excess mortality in the communes and the three arrondissements, compared with the other three atlases of the “SMM_10000”, “DENSI” and “Khi2LnFr” variables. (4) As for the atlas on the SMM_10000 variable, this indicator of excess mortality per 10000 inhabitants helps highlighting differences in small towns, which make up the majority of the French territory. The design limitations are as follows (1) The geomatic processing flows (DB management, statistics, variable formatting, etc.) required to obtain ready-to-use .geojson layers in uMap are cumbersome, complex and not currently reproducible in a single environment. Indeed, the main problem is that our architecture resembles a pipeline system of non-interconnected data, requiring a substantial number of geomatic software tools and several workflows for processing tasks (of which only 2 workflows are fully automated), as in the example of the QGis modeler (see sections 3.2.5: Geomatic data engineering, 3.4.1.1.: step 1: data pipeline description). (2) The legend management system for choropleth maps in color ranges in the uMap platform is not adapted and would require specific development in uMap. (3) Finally, hosting double legends in .png or .jpg format requires another remote server (private and university). This is a constraint of the uMap dashboard, that doesn't anymore allow map legends to be imported in the same place as geojson files. 5.4.3. Discussion on the use of the uMap collaborative platform Benefits of using uMap (1) We have diverted uMap's primary function, which is considered a geomatic environment for collaborative map sharing, to use uMap as a support for our mortality atlases [59]. Indeed, the publication of thematic atlases using choropleth maps is unusual in uMap. There are many public examples of shared maps on various themes (sports routes, itineraries with hotspot symbolism etc.). (2) The France-wide display of the “ FeuTricolore” (TrafficLight) atlas in uMap is highly satisfactory, as it allows instant interpretation of excess mortality alerts. (3) We enhance the appeal of maps with a fairly fine communal grid, so that users can analyze in detail the areas where they live and work (one or more communes). (4) We have not encountered any technical problems displaying instantiated atlases in different web browsers. (5) There is no security maintenance on the uMapOSM.fr server hosting the geojson layers. (6) The possibility of creating a parameter template is offered by map cloning and has been used. Main limitations in terms of use (1) Atlases cannot be grouped within one single uMap URL . (2) An incident with the uMapOSM server at the end of May 2023 caused access errors to all our atlases. This problem was reported on the OpenStreetMap Project forums and GitHub [99] and resolved fairly quickly. 5.5. Discussion based on the example of workflow optimization for double captions and the limits encountered (step 2) We have succeeded in optimizing certain workflows aimed at obtaining . geojson layers in QGis (step 1) and at creating and managing double legends (step 2). Here we describe the optimization of double legends and the limitations encountered. We have succeeded in lightening some of the geomatic processing involved in obtaining double legends, thanks to the .atn scripts (proprietary Adobe Photoshop© (PS7) format) that make up our workflow models. The .atn scripts are obtained from our old PS7 version by recording macros. It's a semi-automated procedure that speeds up the processing of captions, while checking that their formatting, such as resized size, is displayed. In consequences, the . atn script enabled us to save considerable time in processing double captions without and with QRcode. In fact, we avoided manual input errors on multiple, repetitive tasks. Ideally, we would like to work with a more recent version of Photoshop, or move on to other non-proprietary, or free, under certain conditions, drawing software. Concerning the limitations encountered with Photoshop 07 (PS7) (1) . atn scripts can be further optimized using batch processing, currently impossible to realize without bug in the PS7. (2) We noted a problem of incomplete macro recording using the mouse or Interaction Homme Machine ( HMI/HMI, Human Machine Interaction), i.e. the actions recorded in the .atn file. To get around this problem, editing . atn scripts in .xml formatusing existing JavaScript utilities would be a solution worth testing. This would require working on a virtual machine with a suitable older Windows XP operating system for existing utilities [115]. 5.6. Discussion on the proliferation of geomatics tools, languages and tips Software evolution is rapid in IT and geomatics: it is advisable to carry out an organic (dynamic) watch at the beginning and throughout the project [116]. All Computer Aided Mapping and Drawing CAO-DAO/CAMD software used in geomatics, as well as webmapping environments such as collaborative and non-collaborative platforms, require in-depth investment (during the project, with the help of user guides) [124, 148]), a dedicated geomatics forum, and exchanges with IT specialists. These environments generally combine programming languages (Python for scripting in QGis), scripting languages (Scheme) [117] and Visual Basic for Application (VBA) and Philcarto macro languages. Generally speaking, we use the basic functionalities available in geomatics environments as well as utilities made available specifically for an environment such as QGis. For example, for task automation in QGis, one could use the "modeler" tool and its associated .model3 file format for building workflow models. Regarding these languages (Python, Scheme, JavaScript etc.) and open, free, or proprietary tools for automating tasks or interconnecting systems With the aim of (1) optimizing repetitive tasks as much as possible in the course of our needs and projects, or (2) interconnecting systems in a data pipeline [147]: a good compromise for the geomatician is to invest in these geomatic-computing languages and tools in parallel and on an ongoing basis. An intellectual investment is clearly necessary in order to avoid the waste of time and energy caused by the tedious manual execution of repetitive tasks. 5.7. Discussion of atlas results 5.7.1. Atlas "Feu Tricolore" (TrafficLight) semantic relevance and scientific openness Traditionally, maps can be difficult to read for people lacking the prerequisites of thematic cartography, or the concepts of graphic semiology. Our atlas has taken this difficulty into account, offering maps aimed at the general public, healthcare decision-makers and political decision-makers who need relevant, more global information that can be communicated to all. The principle of intuitive comprehension is essential if we are to capture and hold users' attention. In particular, among the instantiated atlases, the one presenting the " FeuTricolore " (TrafficLight) variable is clear for all to see, as it provides an effective account of the given situation for each commune in the national territory. 5.7.2. Atlas Feu Tricolore (TrafficLight)- its warning levels Even though this project is an epidemiological research endeavor, the results of excess mortality in 2020 compared with the two reference years (2018 and 2019) will enable us to examine the causes of excess mortality in certain communes and under mortality in others. In fact, this is one of the interests of our work: to provide this data at a very fine mesh (the commune) and on the scale of the whole of France, including the OverseasDepartmentsandRegions DROMs/OSDRs , enabling each commune to position itself in relation to its region, department or other administrative entity, or in relation to the national level (as in this article). 5.7.3. Indirect epidemiological approach: mapped communal grid and INSEE database There are two epidemiological aspects of interest that emerge indirectly from our geomatic environment of all-cause mortality atlases for the year 2020. The first innovative aspect of our study is the availability of statistical information on deaths from all causes at the municipal level, which by definition is less biased than the one obtained with a more aggregated grid (see section. 2.1.2.1: Cartographic information bias…), and which complements the aggregated cartographic information published on the covid-19 pandemic [1, 37, 149, 150]. The commune is the smallest spatial unit for which INSEE death data are available, with the exception of large cities (Marseille, Lyon and Paris), for which information is available at the arrondissement level. Two additional limitations should also be noted, one relating to discretization thresholds and the other to singular cases (small communal population), either a bias in mortality figures . The second aspect concerns the interest of the source of the death data itself, and our approach, like that realized of French TV channel, France 3 in the Nord - Pas-de-Calais region, of comparing mortality with previous years and with the communal grid. France 3 stated: “Even if the cause of death is not mentioned in the INSEE data, an excess of mortality, compared with the two previous years, can be an indication of the impact of the coronavirus, whose information h has been aggregated at departmental level. This is how we were able to measure, for example, the impact of the deadly heatwave in France in the summer of 2003 [50]”. [9] Support on Github, on the Gitter forum which meets once a month (last Thursday of the month) and on the OSM forum for uMap [10] Publication of the atlas, which is accessible to all, requires the atlas URL to be made available. 6. Perspectives We are to keep in mind that reproducibility of research remains our guideline also for our optimization, implementation and development perspectives for atlases in uMap [121, 151]. 6.1. Finalization of atlases for four periods instead of one in a 2 nd future version In the short term, we intend to supplement our four atlases, including the " FeuTricolore " (TrafficLight) atlas (from the first containment of 2020), with those of a 2 nd version for each of the four periods and for the entire year 2020. The periods, selected as a result of our 2020 watch, correspond to the following four political periods: from January 1st to February 28th inclusive (excluding the covid-19 crisis); from March 1st to May 15th inclusive (1st containment covid-19 crisis); from May 16th to October 28th inclusive (excluding the covid-19 crisis containment); and from October 29th to December 31st inclusive (2nd light containment covid-19 crisis). Four periods would give a better analysis of the phenomenon - i.e. a reflection of the policy over the allotted time. It should be remembered that in this article we present only the " FeuTricolore " (TrafficLight) atlas for the first containment period of 2020. We have initiated a first statistical exploration on a new extraction of death data from INSEE on July 23rd, 2021, and have noticed that the methods used to collect deaths differ between our current version (see Material section 3.1.) and a 2nd version planned over four periods. We already know from the statistics for the 2 nd version that 13944 communesout of 34,833[11] (i.e. 40% of communes) are concerned by an “excess mortality alert” using our “ FeuTricolore” (TrafficLight) variable (2 nd version – 1 st period in the year 2020). Reminder the 1 st version, “see section 4.1.1.: Interpreting the TrafficLight choropleth map”. Secondly, we also plan to deepen the analysis by coupling the atlas results of the four variables and periods with the health policies jointly implemented on the French national territory. This would enable us to draw up a balance sheet of the efficient policy measures taken (in terms of mortality decline during the year 2020) and those with a more nuanced or questionable impact (in terms of non-significant growth and decline). To improve our understanding of the communal landscape, following alerts of excess- or under mortality, we could consider cross-referencing the data with the communes' socio-demographic data. It might also be useful to compare results from France with those from a neighboring country such as Belgium; we are planning to collaborate with a Belgian team, but there are other methodological obstacles to consider. Our different time periods (containment and de-containment) do not strictly correspond; in France and Belgium, for example, the periods (P0, P1, P2 and P3) are partly out of sync [152]. Similarly, it is not certain that in Belgium, the years 2018 and 2019 are reference years, and this should be heeded in the comparative process. Finally, despite some differences between France and Belgium, the methodology applied in this article is transferable to any geographical area for which mortality data (all causes combined) are available. 6.2. Geomatics: focus on optimization, automation and reproducibility in R At the geomatics level, intra- and extra-webmapping processing chains (respectively map production (. geojson layers ) for the “ FeuTricolore” (TrafficLight) atlas for uMap), and "legends and tooltips" information in uMap need to be optimized. We also intend to optimize processing between various environments to interconnect them. To achieve this, we could call on (1) an Extraire, Transformer et Charger ( ETL, Extract, Transform, Load) pipeline architecture from proprietary software, under various types of license; (2) a programming language such as Python, where libraries exist for most of the set objectives [147, 153 ]. Our short- to medium-term goal is to rapidly produce or update atlases instantiated in uMap. In the interests of greater reproducibility, we would also like to focus on a single environment for much of our geomatics processing. To this end, we would like to focus more on the R project environment and its new packages dedicated to cartography[12], i.e. to integrate the entire cartographic processing process rather than working in Excel, Philcarto, Google Earth, VScode and QGis. It is a challenge we believe is achievable: thanks to digital evolution and the interest of developers in geomatics, who are simplifying the work of the humanities and social sciences by simplifying the geomatics processing chain and developing new mapping packages that comply with the rules of the art of graphic semiology. These developers invite us to use the same software environment, and to take advantage of the R project to carry out all data processing chains right through to cartographic results. In this respect, these developers [123, 151] invite us to join the concept of reproducible cartography. Thus, the "Cartography" package, in its new version renamed "mapsf" more user-friendly, lighter and more robust [123, 137, 154] enables all four methodological steps (data collection; data cleaning and ordering; data analysis and spatial representation ) to be followed and carried out within the same software environment to produce maps, and avoids any dispersal or complication of tasks. In other words, it enables a unified workflow and map reproducibility process. The packaged R project is a very good choice for step 1 statistics and geomatics without web-mapping NB: Reproducible cartography raises a stumbling block in the face of sensitive, regulated or even inaccessible health data. Indeed, we could not make health data available for any cartographic reproduction by others, even for a designated community belonging to the healthcare environment. This first barrier is reinforced for a non-healthcare community. Nevertheless, new positive perspectives are emerging thanks to "think tanks" made up of healthcare professionals [155] who are working to improve the reusability of healthcare data, particularly hospital data, for public health and epidemiological research. 6.3. Targeted conceptual and geomatic developments uMap Conceptual evolution: to give more meaning to our atlases like " FeuTricolore " (TrafficLight), for which the uMap platform is originally synonymous with a geomatics environment for collaborative map sharing, we want to share our atlases with the uMap community, with rights open to user interaction. These atlases of all-cause mortality will also be open to personalized comments [59, 133]. We hope to be able to evaluate the interactions and their impact on the attractiveness of the atlases. All-cause mortality atlases at the local level, fed by an open INSEE data source, can join this collaborative map-sharing concept without constraint. Geomatics evolution: uMap's designer has given us the option of installing and developing a customized uMap environment as a basis for implementing various "health atlases". “uMap” would then be implemented on a university server where our atlas legends are currently hosted. During the implementation phase, uMap contributors will be able to contribute to the development of a public[13] fork for the creation and management of legends and the display of tooltips [99, 138]. Beforehand, we plan to present the schematics of (Figure 3 and Figure 7) so that contributors can better assess the complexity of the data pipelines of the two geomatic steps and consider an improvement for the creation of legends in uMap. In addition, this " uMap santé " (uMap in health) could be opened up to our partners and/or other scientists. However, it should be reminded that health data is regulated differently from INSEE data. Within this constrained framework of access and respect for the use of health data, the development of this "uMap santé " (uMap in health) server would be legitimate, but subject to the regulations in force. We could also draw inspiration from the Pixacare medical photo library, with a view to developing and diversifying the use of " uMap santé " (uMap in health) or collaborative health map libraries [156]. [11] Six communes are excluded from the “ FeuTricolore” (TrafficLight) variable, as their population is zero. They are identified by their local INSEE codes: 55039-Beaumont-en-Verdunois; 55050-Bezonvaux; 55139-Cumières-le-Mort-Homme; 55189-Fleury-devant-Douaumont; 55239-Haumont-près-Samogneux and 55307-Louvemont-Côte-du-Poivre. [12] The oldest spatial packages are represented by: RGDAL: interface between R and GDAL libraries; sp: classes and methods for geolocalized data; rgeos: spatial calculations of the following types: area, buffer zone, intersection, superposition, merge, dissolve. The sf package encompasses the 3 previous packages and provides new, simpler-to-use functionalities, as well as being pipe and operator compatible. [13] Software created or modified from existing source code. 7. Conclusions We have demonstrated the value of the uMap Gratuit/Libre et Logiciel de Source Ouverte (GLLSO/FLOSS , free/libre and open-source software) platform [ 59 ] collaborative, interoperable mapping platform in which we instantiated our thematic atlases with choropleth maps and their legends. Our mortality atlas showcase represents good potential for attractiveness and use, thanks to its strengths in terms of - display and visualization of results - ergonomics (usability) and in terms of - understanding and semantic interpretation of cartographic and statistical results. To achieve the desired atlas results, the uMap platform did not require any additional IT development or management of the remote uMap server (handled by OSM.fr ). The standards described in the material section are used in our work. The " FeuTricolore " (TrafficLight) atlas for France appears to be highly relevant in terms of the instant message it communicates, and in this respect, we have provided a complement to the existing covid 19 pandemic cartography that is easy to interpret for any audience. To join the collaborative platform concept, our mortality atlases will be open to sharing since mortality data sourced by INSEE is open and not subject to health data regulations like the Règlement Général sur la Protection des Données (RGPD/GDPR , general data protection regulation). We expect quantitative and critical feedback from users of the uMap environment. Also, the map search by keyword, will enable everyone to access our mortality atlases. In terms of geomatics, since the two-step action plan “without and with webmapping” is rather complex, it would be advisable to consider a number of improvements (reduce the number of software packages used, plan interconnections between software packages to automate processing chains, give priority to geomatic processing in step 1, if possible, only in the R environment). And for step 2, development should be planned with the help of uMap contributors to facilitate automation of the creation of double legends for choropleth maps (in color ranges). At the epidemiological level, our mortality atlases can prove extremely useful for health crisis management if they are developed in real time in uMap or, as in our case, in a retrospective framework on a collaborative platform. Finally, to improve our understanding of the municipal landscape, following the detection of alerts of excess- or under mortality, we could consider coupling our results with socio-demographic data from the municipalities. Declarations Acknowledgements We would like to thank the creators and contributors to the uMap platform, especially Yohan BONIFACE, as well as the geomatics world in particular: the OSGeo and OSGeo.fr geospatial open-source foundations and the Open Street Map association. We would also like to thank all INSEE 's open data suppliers, notably data-gouv, as well as all the developers who make their software available free of charge. Last but not least, we would like to extend our special thanks to our healthcare establishments, who through our activities contribute to public health research and the development of geomatics in healthcare. The authors would like to thank Frank DUFOUR, of the Faculté de Médecine de Nice, RETINES laboratory, for his help in revising the English version of the manuscript. Funding Translation and publication costs are funded by the Groupement des Hôpitaux de l'Institut Catholique de Lille (GHICL , Group of hospitals of the Institut Catholique de Lille). Ms. Quesnel's research activity is supported by the Lille University Hospital. The RETINES laboratory has made available bibliographic referencing software. Availability of data The gross INSEE death dataset extracted May 22, 2020 supporting the conclusions of this article is available in the following healthDataset repository , in [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip]. Author information Authors details 1 Lille University Hospital (CHU de Lille), Regional House of Clinical Research (MRRC), Public Health, Lille, France. Anne QUESNEL-BARBET (AQB) , [email protected] ; [email protected] , https://orcid.org/0000-0003-1038-7344 2 TITSOFT, Montpellier, France. Thierry PAGES (TP) , [email protected] ; https://orcid.org/0000-0002-6075-7633 3 CERIM Lab - EA 2694: Public Health, Faculty of Medicine, University of Lille, Lille, Nord, France. Julien SOULA (JS) [email protected] ; https://orcid.org/0000-0003-0875-7713 4 RETINES Laboratory - Faculty of Medicine, University of Côte d'Azur (UCA), Nice, France, Gilles MAIGNANT (GM) , [email protected] ; https://orcid.org/0000-0003-2017-3972 5 DSI/CIO/CMIO - Catholic Hospitals of Lille (GHICL), Nord, Lille, France. Arnaud HANSSKE (AH) , [email protected] ; https://orcid.org/0000-0001-7029-2318 Corresponding author Anne QUESNEL-BARBET, E-mail address: [email protected] ; [email protected] CHU de Lille, Maison Régionale de la Recherche Clinique, Santé Publique, Lille, France. Authors' contributions Study design: AQB. Data preparation: TP, AQB. Server management: JS Study and statistical analysis: AQB, TP, GM. Drafting and finalization of manuscript: AQB, GM, TP, AH. All authors have read and approved the final manuscript. Declarations Ethics approval and consent to participate This article does not contain any studies with human participants or animals performed by any of the authors. Statement on conflicts of interest All authors declare that they have no conflict of interest. Consent for publication Not applicable References Fry R, Hollinghurst J, Stagg HR, Thompson DA, Fronterre C, Orton C, Lyons RA, Ford DV, Sheikh A, Diggle PJ: Real-time spatial health surveillance: Mapping the UK COVID-19 epidemic . Int J Med Inform 2021, 149 (104400):1-8. https://doi.org/10.1016/j.ijmedinf.2021.104400. Web mapping or online mapping. [https://en.wikipedia.org/wiki/Web_mapping]. Accessed 17 July 2024. Brunet R: Map and sketch: report on Arnould C, Denisart F, Mangin C, 1999, "Geography sketches for the baccalauréat", Paris: Ellipses. And on: Poidevin D, 1999, "The Action Map. A practical guide to map design and production", Paris: Ellipses. Mappemonde 2000, 60 (4):45-46. Poidevin D: The Map as a Means of Action. A practical guide to card design and production / La Carte moyen d’action. Guide pratique pour la conception et la réalisation de cartes . Paris, France: Ellipses; 1999. Ravenea u J: Brunet R, (1987) The map, instructions for use / La carte, mode d’emploi. Paris, Fayard/Reclus, 270 p., 269 ill. Cahiers de géographie du Québec 1988, 32 (85):99–102. https://doi.org/10.7202/021949ar. Quesnel-Barbet A, Dufossez F, Souris M, Delerue D, Hansske A: Geomatics in health and applications in the north of France . Techniques de l'ingénieur: Électronique, robotique et TIC pour la santé 2016, TIB 628 (med4000):5; 1-41. https://doi.org/10.51257/a-v1-med4000. The Home of Location Technology Innovation and Collaboration. [https://www.ogc.org/standards/]. Accessed 17 July 2024. FAIR data - FAIR principles, as published by GO FAIR - Findable, Accessible, Interoperable and Reusable (FAIR). [https://en.wikipedia.org/wiki/FAIR_data]. Accessed 17 July 2024. Hayat. Flora: Production of digital commons and cartographic use . Paris, France: Paris Cité; 2019.[http://www.theses.fr/2019UNIP7135/document]. Accessed 17 July 2024. Welcome to the OpenStreetMap wiki. [https://wiki.openstreetmap.org/wiki/Main_Page]. Accessed 17 July 2024. OpenStreetMap (OSM) [https://en.wikipedia.org/wiki/OpenStreetMap]. Accessed 17 July 2024. Open data: the opening up of public data - Free software issues, open standards and interoperability / L'open data: l'ouverture des données publiques - Enjeux du logiciel libre, standards ouverts et interopérabilité. [http://www.apitux.org/index.php?2010/11/19/235-opendata-l-ouverture-des-donnees-publiques]. Accessed 17 July 2024. BECQUET J-C: Open GIS, software and data conference: technical training for students, professionals and jobseekers / Colloque SIG libres, logiciels et données: des formations techniques pour étudiants, professionnels et demandeurs d'emploi . In: Le monde du libre: le logiciel libre et les données ouvertes au service des territoires: 30 juin 2006; XSALTO, Digne-les-Bains, France. : APITUX; 2006. [https://www.agendadulibre.org/events/587; http://www.apitux.org/index.php?q=COLLOQUE]. Accessed 17 July 2024. Create interactive and crowd-sourced web maps. [https://about.canvis.app/]. Accessed 17 July 2024. Crowdsourcing - Participatory production - Open outsourcing. [https://en.wikipedia.org/wiki/Crowdsourcing]. Accessed 17 July 2024. Crowdmapping. [https://en.wikipedia.org/wiki/Crowdmapping]. Accessed 17 July 2024. Noucher M, Gautreau P: Is open access reshuffling the deck? New perspectives for geographical data / Le libre accès rebat-il les cartes? Nouvelles perspectives pour les données géographiques. Les Cahiers du numérique 2013, 9 (1):57-83. Open source. [https://en.wikipedia.org/wiki/Open_source]. Accessed 17 July 2024. Web 2.0, sometimes called participative Web. [https://en.wikipedia.org/wiki/Web_2.0]. Accessed 17 July 2024. BECQUET J-C: Open formats and interoperability / Formats ouverts et interopérabilité . 2004, 84 (84).[http://www.apitux.org/index.php?2008/07/25/84-formats-ouverts-et-interoperabilite]. Accessed 17 July 2024. Interoperability. [https://en.wikipedia.org/wiki/Interoperability]. Accessed 17 July 2024. What is a data pipeline? [https://www.ibm.com/topics/data-pipeline]. Accessed 17 July 2024. What is a data pipeline and how does it work? Definition of a data pipeline / Qu’est-ce qu’un data pipeline et comment cela fonctionne? Définition d'un pipeline de données. [https://www.talend.com/fr/resources/what-is-a-data-pipeline/#:~:text=Un%20pipeline%20de%20donn%C3%A9es%20englobe,donn%C3%A9es%20pr%C3%AAtes%20%C3%A0%20%C3%AAtre%20exploit%C3%A9es]. Accessed 17 July 2024. BERNARD M, BRIONE M: GIS Observatory: Norms and standards, quo vadis? / Observatoire du SIG: Normes et standards, quo vadis? In: Observatoire du SIG. Edited by CNIG. Paris, France: SIGMAG TV.fr; 2022: 1-10. [https://cnig.gouv.fr/IMG/pdf/_c_vpw_sigmag_sigtv.fr_n34_obervatoire_du_sig_enquete_normes_standards_-_octobre_2022.pdf]. Accessed 17 July 2024. French Association for Standardization (AFNOR.org) in cooperation with the International Organization for Standardization (ISO.org) [https://www.iso.org/organization/1738.html]. Accessed 17 July 2024. Open Geospatial Consortium (OGC), Inc. [https://www.iso.org/organization/9825.html]. Accessed 17 July 2024. Etalab: a French public administration to improve public service and public action through data / Etalab: une administration française pour améliorer le service public et l'action publique grâce aux données. [https://fr.wikipedia.org/wiki/Etalab]. Accessed 17 July 2024. Data Act: Proposal For A Regulation On Harmonised Rules On Fair Access To And Use Of Data. [https://digital-strategy.ec.europa.eu/en/library/data-act-proposal-regulation-harmonised-rules-fair-access-and-use-data]. Accessed 17 July 2024. Geodetic parameter data set EPSG. [https://en.wikipedia.org/wiki/EPSG_Geodetic_Parameter_Dataset#:~:text=EPSG%3A4326%20%2D%20WGS%2084%2C,including%20Google%20Maps%20and%20OpenStreetMap]. Accessed 17 July 2024. ISO 19111:2019 Geographic information — Coordinate reference system. [https://www.iso.org/standard/74039.html]. Accessed 17 July 2024. EPSG:4326 -- WGS84 - World Geodetic System 1984, used in GPS. [https://epsg.io/4326]. Accessed 17 July 2024. IT applications in science; Including digital geographic information. [https://www.iso.org/ics/35.240.70/x/]. Accessed 17 July 2024. Developing With Web Standards – Recommendations and best practices. [https://www.456bereastreet.com/lab/developing_with_web_standards/]. Accessed 17 July 2024. Sarafinof D, Mistre A, Picinbono G, Vallet B, Heydel L: The GeoBIM approach: from land management to building management / La démarche GéoBIM: de la gestion du territoire à celle d’un bâtiment . Annales des Mines - Responsabilité et environnement 2019, 94 (2):42-46. https://doi.org/10.3917/re1.094.0042. CityGML GeoBIM ADE - This GeoBIM standard was launched in 2009 . In . : [https://www.citygmlwiki.org/index.php?title=Citygml_Wiki]; 2009. Accessed 17 July 2024. DGIWG: The Defence Geospatial Information Working Group. [https://dgiwg.org/about]. Accessed 17 July 2024. Scancovid. [https://scancovid.scansante.fr/scancovid/_w_70c5f626/?code=open_data]. Accessed 17 July 2024. Care and territories / Soins-et-territoires. [https://www.scansante.fr/actualites/soins-et-territoires]. Accessed 17 July 2024. Cromley EK, McLafferty SL: GIS and Public Health : Guilford Press; 2011. Jones K, Duncan C, Moon G: Context, scale, and the geography of health: understanding the relevance of ecological studies . American Journal of Public Health 1998, 88 (2):216-222. Kwan MP: The uncertain geographic context problem . Annals of the Association of American Geographers 2013, 103 (2):221-244. https://doi.org/10.1080/00045608.2012.687349. Kazi AM, Ali M, Ayub K, Kalimuddin H, Zubair K, Kazi AN, Artani A, Ali SA: Geo-spatial reporting for monitoring of household immunization coverage through mobile phones: Findings from a feasibility study . Int J Med Inform 2017, 107 :48-55. https://doi.org/10.1016/j.ijmedinf.2017.09.004. Plantier M, Havet N, Durand T, Caquot N, Amaz C, Philip I, Biron P, Perrier L: Does adoption of electronic health records improve organizational performances of hospital surgical units? Results from the French e-SI (PREPS-SIPS) study . Int J Med Inform 2017, 98 :47-55. https://doi.org/10.1016/j.ijmedinf.2016.12.002. Alam K, Mahumud RA, Alam F, Keramat SA, Erdiaw-Kwasie MO, Sarker AR: Determinants of access to eHealth services in regional Australia . Int J Med Inform 2019, 131 :103960. https://doi.org/10.1016/j.ijmedinf.2019.103960. Naghavi M, Mohsen N, Contributors.: Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013 . Lancet 2015, 385 (9963):117-171. https://doi.org/10.1016/s0140-6736(14)61682-2. Cancers in France 2015 edition / Les cancers en France édition 2015. [https://www.e-cancer.fr/content/download/148692/1867381/file/Les-cancers-en-France-edition-2015.pdf]. Accessed 17 july 2024. Geodes: Geo Data in Public Health, the cartographic observatory of epidemiological indicators / Geodes: Géo Données En Santé Publique, l’observatoire cartographique des indicateurs épidémiologiques. [https://geodes.santepubliquefrance.fr/#view=map2&c=indicator]. Accessed 17 July 2024.. The health monitoring institute / L'institut de veille sanitaire. [https://www.has-sante.fr/jcms/c_436647/fr/institut-de-veille-sanitaire#:~:text=L'Institut%20de%20veille%20sanitaire,la%20population%20et%20son%20%C3%A9volution.]. Accessed 17 July 2024.. Sentinelles. [https://www.sentiweb.fr/]. Accessed 17 July 2024. Survey: Coronavirus: which towns in the Nord and Pas-de-Calais departments have recorded higher mortality since 1 March? / Enquête: Coronavirus : quelles communes du Nord et du Pas-de-Calais enregistrent une surmortalité depuis le 1er mars ? [https://france3-regions.francetvinfo.fr/hauts-de-france/enquete-coronavirus-quelles-communes-du-nord-du-pas-calais-enregistrent-surmortalite-1er-mars-1825674.html?fbclid=IwAR0ue9PUte3GmDvyRKI44WHl2An4YTjrbHUl8pFTiSEOmg7gz3BvrkTuQ-M#La%20surmortalit%C3%A9%20dans%20l'arrondissement%20de%20Lille]. Accessed 17 July 2024. WHO.: World health statistics 2014 . In . Edited by World-Health-Organization: Apps. [https://www.who.int/news/item/15-05-2014-world-health-statistics-2014]; 2014. Accessed 17 July 2024. Quesnel-Barbet A, Soula J, Sauleau E-A, Bazile P, Dufossez F, Albert F, Hansske A, Trilling L, Maignant G: PoleSat, health planning modelling for decision support. Application to a reorganization of vascular catheterization provision, at the geographic level of "territorial hospital grouping . ISTE openscience Oct 2022, Numéro spécial: Gestion et ingénierie des systèmes hospitaliers (5):21. https://doi.org/10.21494/ISTE.OP.2022.0883. Quesnel-Barbet A, Soula J, Sauleau E-A, Bazile P, Dufossez F, Maignant G, Albert F, Hansske A, Trilling L: PoleSat, health planning modelling for decision support. Application to a reorganization of vascular catheterization provision, at the geographic level of "territorial hospital grouping ISTE openscience Janv 2022, Numéro spécial: Gestion et ingénierie des systèmes hospitaliers (5):27. https://doi.org/10.21494/ISTE.OP.2022.0774. Quesnel A, Soula J, Wyndels K, Beuscart R: POLESAT: e-Atlas of Health, Planning and Sustainable Development . Ingénierie et recherche biomédicale 2009, 30 (4):201-204. https://doi.org/10.1016/j.irbm.2009.05.005. Quesnel-Barbet A, Soula J, Dufossez F, Sharma N, Ruhela A, Beuscart R: POLESAT, an innovative e-health geomatic platform in decision-making based on: Geographical approach, medical knowledge visualization and geographic information system and web-mapping architecture . IRBM 2013, 34 (4-5):267-273. https://doi.org/10.1016/j.irbm.2013.06.001. Domain name. [https://en.wikipedia.org/wiki/Domain_name]. Accessed 17 July 2024. Review 2018 OSM France services / Revue 2018 Les services OSM France. [https://www.openstreetmap.fr/revue-2018-les-services-osm-france/]. Accessed 27 June 2023. French all-cause mortality monitoring atlas (health crisis - SARS COV-2). Period from 1st March to 15th May inclusive 2020 - 1st containment. Excess mortality alerts expressed by Traffic-Light - uMap. [http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.793/4.570]. Accessed 17 July 2024. Shahamati S, Baumann Y, Caquard S, Denieul-Pinsky L, Shaw E: uMap: A Free, Open-Source Alternative to Google My Maps . CartographicPerspectivesorg 2022(99):6–18. https://doi.org/10.14714/CP99.1729. Gentleman R, Ihaka R, R Core Team., Bates D, Chambers J, Dalgaard P, Gentilhomme R, Hornik K, Isaac R, Kalibera T et al : The R Project. The Comprehensive R Archive Network: R for Windows . V 4.3.1 edn Global community; 1997: GNU GPL v2. [cran.r-project]. Accessed 17 July 2024. QGIS-Development-Team.: QGIS is A Free, Open Source, Cross Platform (lin/win/mac) Geographic Information System (GIS) . V 3.18.1-Zürich edn. Global community: Qgis.org; 2002: CC BY-SA 3.0; GPL-2.0. [github.com/qgis/QGIS]. Accessed 16 July 2024. Digital cartography workshop with uMap / Atelier cartographie numérique avec uMap . [https://associationici.fr/wp-content/uploads/2018/09/Guide-de-prise-en-main-Umap-OSM-v3.pdf]. Accessed 16 July 2024. Gross death file from 2017 to 2020: Daily death files download site - Latest files released - Daily death files by sex, age and place - May 22, 2020 / Fichier de décès brut de 2017 à 2020: Site de téléchargement des fichiers des décès quotidiens - Derniers fichiers diffusés - Fichiers des décès quotidiens par sexe, âge et lieu - 22 mai 2020 . [https://www.insee.fr]; [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip]; https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/deathDataset]; [https://www.data.gouv.fr/fr/datasets/fichier-des-personnes-decedees/#/resources]. Accessed 16 July 2024. Coronavirus: why the errors in counting deaths in nursing homes? / Coronavirus: pourquoi ces erreurs dans les décomptes des décès en Ehpad? [https://www.leparisien.fr/societe/coronavirus-pourquoi-ces-erreurs-dans-les-decomptes-des-deces-en-ehpad-20-05-2020-8320674.php]. Accessed 27 June 2024. SARS-CoV-2, 68 communes have reached the alert threshold - (alert threshold set at 20 new contaminations per week per 100,000 inhabitants) / Covid-19 : en Belgique, 68 communes ont atteint le seuil d'alerte -( seuil d'alerte fixé à 20 nouvelles contaminations par semaine pour 100 000 habitants ). [https://france3-regions.francetvinfo.fr/hauts-de-france/covid-19-belgique-soixante-communes-ont-atteint-seuil-alerte-1856270.html?fbclid=IwAR2eSX_aVs1rPn7qLU_ZiNKPos0_XVl_q4IRmAy5seknZaLFu0AHn93Q2fE#xtor=EPR-521-[france3regions]-20200723-[info-titre1]&pid=726375-1583524212-a50ad1ea]. Accessed 16 July 2024. Covid-19: what the 150 French clusters reveal about the epidemic / Covid-19: ce que révèlent les 150 clusters français de l’épidémie. [https://www.lefigaro.fr/sciences/covid-19-ce-que-revelent-les-150-clusters-francais-de-l-epidemie-20200605?origine=VWT16001&utm_campaign=echobox&utm_medium=Social&utm_source=Facebook&fbclid=IwAR2_RHpbQK7c1Urqlw-TTQJjJ_EuhCKnW3M-X6na6rgQKxt6rheqfJ22NLU#Echobox=1591387766]. Accessed 27 June 2024. Download the 2017 legal population file (excluding Mayotte) / Téléchargement du fichier d'ensemble des populations légales en 2017 (hors Mayotte). [https://www.insee.fr/fr/statistiques/4265429]. Accessed 16 July 2024. Legal population of Mayotte in 2017, Population census / Populations légales de Mayotte en 2017 - Recensement de la Population. [https://www.insee.fr/fr/statistiques/2120838]. Accessed 16 July 2024. ADMIN-EXPRESS-COG 2020 edition by territory: the administrative division of the French territory (commune, departmental district, department, region...) / ADMIN-EXPRESS-COG 2020 édition par territoire: la division administrative du territoire français (commune, arrondissement, département, région...). [https://geoservices.ign.fr/adminexpress]. Accessed 27 June 2024. Pages T, Soula J, Quesnel-Barbet A: Death dataset from 2018-2020 linked to supplementary_file_1 / Jeu de données sur les décès de 2018-2020 liés au fichier supplémentaire_1 . [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/SF1_supplementary_file_1.zip]. Accessed 16 July 2024. Wagniez P: Philcarto utility guide: Eclats2.xxDOC / Guide de l'utilitaire pour Philcarto: Eclats2.xxDOC . 2019 [http://philcarto.free.fr/02_telecharger_logiciels/02_3_telecharger_eclats/fichiers_installer_eclats/Eclats2.xxDOC.zip]. Accessed 16 July 2024. Wagniez P, Contributors.: Philcarto: Eclat utility / Philcarto: utilitaire Eclat . V 2.10.h. France: 2023: freeware. [http://philcarto.free.fr/02_telecharger_logiciels_nouveaux_2023/02_1_telecharger_2023.html]; Accessed 16 July 2024. data.gouv.fr., Lexman A: Map of departments, Metropolitan France and DROM - Open Data Commons Open Database License (ODbL) / Carte des départements, France Métropolitaine et DROM - ODbL . [https://www.data.gouv.fr/fr/datasets/carte-des-departements-2-1/; https://github.com/gregoiredavid/france-geojson/blob/master/departements.geojson]. Accessed 27 June 2024. data.gouv.fr., Lexman A: Map of French regions (new regions in 2016) / Carte des régions françaises (nouvelles régions de 2016) . [https://www.data.gouv.fr/fr/datasets/carte-des-regions-francaises-nouvelles-regions-de-2016/; https://portailsig.org/content/recuperer-des-donnees-openstreetmap-gdalogr.html]. Accessed 16 July 2024. data.gouv.fr., OSM-Contributors.: Location of intensive care units - services-de-reanimation.geojson Open Data Commons Open Database License (ODbL) / Localisation des services hospitaliers de réanimation - services-de-reanimation.geojson, (ODbL) . [https://www.data.gouv.fr/fr/datasets/localisation-des-services-hospitaliers-de-reanimation/]. Accessed 16 July 2024. data.gouv.fr., OSM-Contributors.: Location of emergency services: data - services-daccueil-des-urgences.geojson - Open Data Commons Open Database License (ODbL) / Localisation des services d'accueil des urgences: data - services-daccueil-des-urgences.geojson (ODbL) . [https://www.data.gouv.fr/fr/datasets/localisation-des-services-daccueil-des-urgences/#/resources]. Accessed 16 July 2024. geo.data.gouv.fr: Switching off geo.data.gouv.fr datasets- France - Emergency medicine and authorizations / Extinction des jeux de données sur geo.data.gouv.fr - France - Médecine d'urgence et autorisations . [https://www.data.gouv.fr/fr/posts/extinction-de-geo-data-gouv-fr/]. Accessed 16 July 2024. SQLite: A Database Engine Multiplatform - Public Domain License. [https://en.wikipedia.org/wiki/SQLite]. Accessed 27 June 2024. Hipp D-R: SQLite: Most Widely Deployed and Used Database Engine . V3.42.0 edn. North Carolina, USA; 2000: Open-Source, not Open-Contribution. [sqlite.org]. Accessed 16 July 2024. Help - updating R, Rstudio and packages / Aide - mise à jour de R, Rstudio et packages. [https://delladata.fr/tutoriel-mise-a-jour-de-r/]. Accessed 27 June 2024. Allaire JJ, Gutman J, Mcphers J, Cheng J, De-Vries A, Chen D, Arendt C, Scheidegger C, Falbel D, Aden-Buie G et al : RStudio IDE . V 2023.06.1+524 edn. Global community, 2011: Open Source (AGPL v3). [posit.co/products/open-source/rstudio/]; [github.com/rstudio/rstudio]. Accessed 16 July 2024. Wickham H-A: Plyr library: Tools for Splitting, Applying and Combining Data . V1.8.7 edn. Houston, TX, USA: 2014: Copyright Holder: Hadley Wickham, MIT license. [cran.r-project]; [plyr.had.co.nz]. Accessed 16 July 2024. Allaire J, Xie Y, Dervieux C, McPherson J, Luraschi J, Ushey K, Atkins A, Wickham H, Cheng J, Chang W et al : Rmarkdown: Dynamic Documents for R. R package . V 2.21 edn. Global community: cran.r-project.org; 2007: GPL-3. [github.com/rstudio/rmarkdown]. Accessed 16 July 2024. Walker A, Schauberger P, Braglia L, Sturm J, Garbuszus J-M, Barbone J-M, Contributors.: Openxlsx library: Read, Write and Edit xlsx Files . V 4.2.5 edn. Sidney, Australia; Vienne, Austria 2014: MIT-license. [https://ycphs.github.io/openxlsx/]; [github.com/ycphs/openxlsx]. Accessed 16 July 2024. GDAL: an MIT-licensed open-source translator library for raster and vector geospatial data formats - the May 2023 GDAL/OGR v3.7.0. [https://github.com/OSGeo/]. Accessed 27 June 2024. WANIEZ P: Philcarto . In . , v2021.d edn. Bordeaux, France; 2010: freeware. [Philcarto.free.fr]. Accessed 16 July 2024. Bloch M: Mapshaper: Tools for editing Shapefile, GeoJSON, TopoJSON and CSV files . V0.6.33 edn. NY, USA: 2014: MPL 2.0 license. [mapshaper.org]; [github.com/mbloch/mapshaper/wiki]. Accessed 16 July 2024. MacWright T, Whong C, Contributors.: GEOJSON file editor: geojson.io . In: OSM-mapboxcom - Maxarcom. Edited by rolling (2014-08-11) edn. NY, USA; Global community 2022: ISC License. [geojson.io] ; [github.com/mapbox/geojson.io]. Accessed 16 July 2024. Gougelet P-E, Mezich T, Budz45.: XnView is a general-purpose image organizer and file manager. Proprietary license, V 2.51.2 edn. Reims, France; 1998, [xnview.com/fr/about/]. Accessed 26 June 2024. Gougelet P-E: XnConvert. V 1.98 - Windows x64; v 7.99 Libformat edn. Reims, France: Dec 2021: Proprietary license. [newsgroup.xnview.com/viewforum.php?f=79]. Accessed 26 June 2024. Irfan S: Irfan in: Supports Windows XP, Vista, 7, 8, 10 and 11. V 4.5.9 edn. Austria; 1996-2023: Freeware for non-commercial use. [irfanview.com]. Accessed 26 June 2024. Adobe.: Photoshop . V 7.0 edn. Mountain View, California, USA: 1990-2002: Proprietary license. [Adobe.com]. Accessed 16 July 2024. Microsoft: Paint. V 11.2301.22.0 edn. USA: 2000: Proprietary license. [support.microsoft.com]. Accessed 26 June 2024. dotPDN-LLC., Brewster R, Contributors.: Paint.NET image and photo editing software for Windows PCs. V 5.0.7 edn. Washington, USA; 2019: Classic, free-of-charge. [Getpaint.net]. Accessed 26 June 2024. Visual Basic for Applications. [https://en.wikipedia.org/wiki/Visual_Basic_for_Applications]. Accessed 26 June 2024. Tan V: Open-source developer-friendly spreadsheet library compatible with Microsoft Excel . V 3.5.0 (Sept 2020) edn. Singapore: 2011: MIT License. [spreadsheetlight.com]. Accessed 26 June 2024. Boniface Y: uMap-project and source code . In . Edited by github.com, v 1.4.0 edn. France: 2013: WTFPL license. [github.com/umap-project/umap/]. Accessed 26 June 2024. Do What The Fuck You Want To Public License (WTFPL). [https://en.wikipedia.org/wiki/WTFPL]. Accessed 26 June 2024. Boniface Y: uMap . In . Edited by github.com, v 1.4.0 edn. France; 2013: WTFPL license. [https://umap.openstreetmap.fr/en/]; [https://github.com/umap-project/umap/issues/1115]. Accessed 26 June 2024. uMap. [https://wiki.openstreetmap.org/wiki/UMap]. Accessed 26 June 2024. What's the Difference Between Domain Names and Web Hosting? / Quelle Différence Entre Nom de Domaine et Hébergement Web? [https://www.hostinger.fr/tutoriels/nom-de-domaine-et-hebergement#Quelle_est_la_difference_entre_nom_de_domaine_et_hebergement_web]. Accessed 27 June 2024. Google.: Google Earth For 3D representation of Earth . V 7.3.6.9345 edn. Menlo Park, California, USA: Google.com; 2001: Freeware. [https://earth.google.com/]. Accessed 26 June 2024. Novak T: Geo Data Viewer - ViewMap GDV-4: geographic data viewer In . Edited by [github.com/RandomFractals/geo-data-viewer], v 2.6.0 edn. Chicago, Illinois, USA: [marketplace.visualstudio.com/items?itemName=RandomFractalsInc.geo-data-viewer]; 2020: Apache License, V 2.0. Accessed 17 July 2024. Schindelin J, Xhmikos R, Oliff C, Contributors.: Git Bash For Bash Emulation and Connect In The Secure Shell Protocol (SSH) . In . Edited by github.com, v 2.40.0 edn. Global community: gitforwindows.org; 2014: GNU- GP License v2. [github.com/git-for-windows/git/wiki/FAQ/_history?page=3]. Accessed 26 June 2024. Keyhole Markup Language (KML) : standard of the Open Geospatial Consortium (OGC) [https://en.wikipedia.org/wiki/Keyhole_Markup_Language]. Accessed 26 June 2024. Scalable Vector Graphics (SVG) and Geographic Information System. [https://www.directionsmag.com/article/3916]. Accessed 28 June 2024. W3C.org: Scalable Vector Graphics (SVG) Version 2 features and syntax specification . In: Standard. 2018. [https://github.com/w3c/svgwg]. Accessed 28 June 2024. Butler H, Daly M, Doyle A, Gillies S, Hagen S, Schaub T: RFC 7946 - The GeoJSON Format, august 2016 - proposed standard . In [rfc-editor.org]; 2016. https://doi.org/10.17487/RFC7946. The GeoJSON Specification (RFC 7946) - a format for encoding a variety of geographic data structures. [https://geojson.org/]. Accessed 28 June 2024. Geographic JSON (geojson) . In: Standard. IETF; 2022. [Datatracker.ietf.org]. Accessed 28 June 2024. Shapefile, proprietary format now opened by ESRI as a 'standard' with extensions .shp, .shx, .dbf. [https://fr.wikipedia.org/wiki/Shapefile]. Accessed 28 June 2024. Shapefile or Geopackage: which format to choose? / Shapefile ou Geopackage: quel format choisir? [https://geoma-sig.com/2020/09/29/shapefile-ou-geopackage-quel-format-choisir/]. Accessed 27 June 2024. Geospatial Data and GeoJSON and Draw and Edit with GeoJson.io. [https://handsondataviz.org/geojson.html; https://handsondataviz.org/geojsonio.html]. Accessed 28 June 2024. 17.17. Getting Started with the QGIS' Graphical Modeler. [https://docs.qgis.org/3.28/fr/docs/training_manual/processing/modeler_twi.html]. Accessed 28 June 2024. How do I convert Photoshop ATN files to XML? / Comment faire pour convertir des fichiers Photoshop ATN à XML? [https://community.adobe.com/t5/photoshop-ecosystem-discussions/convert-xml-to-atn-action-file/m-p/11935514]. Accessed 16 July 2024. How do I monitor my computer? / Comment je fais ma veille? [https://grafikart.fr/blog/veille-grafikart-organique]. Accessed 14 July 2024. Kimball S, Mattis P, Mena F, Singh M, Neumann S, Natterer M, Neary D, Nordholts M, Jehan.: GIMP: The Free & Open-Source Image Editor . V 2.10.34 edn. Berkeley, California, USA: Gimp.org; 1995: CC BY-SA 4.0 License. [Gitlab.gnome.org]. Accessed 17 July 2024. Andler JA, Bintz J, Marquardt C, Böck H, Urošević A, Engelen J, Kiirala N, Taxel P, Hetherington C, Conde-Rueda X et al : Inkscape . V 1.2.2 edn. Global community: Inkscape.org; 2003: GPLv2 License. [github.com/inkscape/inkscape]. Accessed 17 July 2024. Simple Inkscape Scripting For Inkscape. [https://github.com/spakin/SimpInkScr]. Accessed 28 June 2024. Bogle B, Gundersen E: Mapbox: a collaborative platform . V2.15.0 edn. San Francisco, Californie, USA: Mapbox.com; 2010: Proprietary license. [github.com/mapbox]. Accessed 17 July 2024. Tuto@Mate (SHS - Human Social Science) Thimotee Giraud Package R Cartography / Tuto@Mate (SHS - Sciences humaines et sociales) Thimotee Giraud Package R Cartographie. [https://www.youtube.com/watch?v=OI3_AOg6pfc&t=129s]. Accessed 26 June 2024. Giraud T, Hernangómez D, Pecout H, Ysebaert R, Fellows I, Lemon J, Zenoni F: Mapsf: Thematic Cartography. R package . In: Magrit, cartography. Edited by [github.com/riatelab/mapsf/], v 0.7.0 edn. Paris, France: 2021: GPL-3. [riatelab.github.io/mapsf/]. Accessed 17 July 2024. Giraud T, Lambert N: R package: Cartography . V 3.1.3 edn. Paris, France: cloud.r-project.org; 2017: GPL-3. [github.com/riatelab/cartography]. Accessed 17 July 2024. Wagniez P: Phicarto guide: Philcarto_2020_MAJ_DOC / Guide de Phicarto: Philcarto_2020_MAJ_DOC ; 2019:[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/Philcarto_2020_MAJ_DOC.zip]. Accessed 28 June 2024. Wagniez P: Philcarto's guide: DOCdeGRANITn1 / Guide de Philcarto: DOCdeGRANITn1 ; 2023.[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/DOCdeGRANITn1.zip]. Accessed 28 June 2024. Wagniez P: Philcarto_2020_QuickStartGuide in 11 lessons for beginners / Philcarto_2020_GuideDemarrageRapide en 11 leçons pour les débutants ; 2020:[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/PhilcartoGuideDemarrage.zip]. Accessed 28 June 2024. Jenks Natural Breaks — The Best Range Finder algorithm. [https://medium.com/analytics-vidhya/jenks-natural-breaks-best-range-finder-algorithm-8d1907192051]. Accessed 28 June 2024. Jenks natural breaks optimization. [https://en.wikipedia.org/wiki/Jenks_natural_breaks_optimization]. Accessed 26 June 2024. Edit and Join with Mapshaper. [https://handsondataviz.org/mapshaper.html#]. Accessed 28 June 2024. Protection of map data - uMap / Protection des données cartographiques uMap. [https://forum.openstreetmap.fr/t/protection-des-donnees-cartographiques-umap/7251]. Accessed 28 June 2024. Tutorials: creating a personalized map with uMap / Tutoriels: créer une carte personnalisée avec uMap. [https://wiki.cartocite.fr/doku.php?id=umap:tutoriel_umap]. Accessed 28 June 2024. uMap" user guide / Guide d'utilisation "uMap" . [https://framalibre.org/content/umap]. Accessed 28 June 2024. Imagined Forrest Gump run (1994 film). uMap: exemple de diaporama [http://umap.openstreetmap.fr/en/map/imagined-forrest-gump-run-1994-film_117755#5/39.113/-93.384]. Accessed 26 June 2024. MyOSMatic - Online web printing service / Service web d'impression en ligne. [https://print.get-map.org/]. Accessed 14 July 2024. Additional census documentation / Documentation complémentaire sur le recensement. [https://www.insee.fr/fr/information/2383410]. Accessed 27 June 2024. Understanding Legal Populations / Comprendre les populations légales. [https://www.insee.fr/fr/information/2553979]. Accessed 27 June 2024. Thematic Maps With Mapsf (R Package). [https://riatelab.github.io/mapsf/]. Accessed 26 June 2024. uMap developer documentation -installation on ubuntu - linux. [https://umap-project.readthedocs.io/en/master/]. Accessed 28 June 2024. Django project documentation: Web framework [https://docs.djangoproject.com/en/4.2/]. Accessed 14 July 2024. Boniface Y: umap-project: Add An API For Automatic Export . In: /umap/issues/78. v 1.4.0. edn. France; 2014: WTFPL license. [github.com/umap-project/]. Accessed 28 June 2024. L'API MapOsmatic: uMap GeoJson Export API Support / Prise en charge de l'API MapOsmatic d'exportation Umap GeoJson. [https://blog.get-map.org/index.php/2019/06/16/umap-geojson-export-api-support/]. Accessed 14 July 2024. Forum uMap. [https://forum.openstreetmap.fr/c/utiliser/umap/29]. Accessed 14 July 2024. Community gitter forum for the umap-project / umap. [https://app.gitter.im/#/room/#umap-project_umap:gitter.im]. Accessed 14 July 2024. Map search on the keyword "health" hosted on the OSM server and shared with the community. [https://umap.openstreetmap.fr/fr/search/?q=health]. Accessed 26 June 2024. GPX file format (v 1.1): Open Geospatial Consortium (OGC)' standard. [https://en.wikipedia.org/wiki/GPS_Exchange_Format]. Accessed 26 June 2024. Importing remote data from a CSV file - Using / Importer des données distantes depuis un fichier CSV [https://forum.openstreetmap.fr/t/importer-des-donnees-distantes-depuis-un-fichier-csv/2601/3]. Accessed 14 July 2024. I built a geospatial ETL pipeline with python and this is what I learned. [https://medium.com/analytics-vidhya/i-built-a-geospatial-etl-from-scratch-with-python-and-this-is-what-i-learned-b45b37d15f94]. Accessed 14 July 2024. Photoshoplus offers settings, tutorials and colors for Photoshop. And it's all free! [https://www.photoshoplus.fr/]. Accessed 14 July 2024. ROZIER Guillaume: Covidtracker.fr dashboard several analysis scales . In . Edited by github.com. Paris, France: OVHCloud; 2020: MIT license. [https://github.com/rozierguillaume/covidtracker-tools]. Accessed 14 July 2024. Fabacher T: Covid-19 outbreak - dynamic atlas, global and national scales / Covid-19 outbreak - atlas dynamique, échelles mondiale et nationale . Strasbourg, France; 2020: GMRC-ICUBE-Lab Freeware. [https://teststid.shinyapps.io/covid_19/]. Accessed 17 July 2024. Giraud T, Lambert N: Reproducible Cartography . In: Advances in Cartography and GIScience: 2017 : Springer Cham; 2017: 173-183. https://doi.org/10.1007/978-3-319-57336-6. One year of confinement: some dates that marked the lives of Belgians (chronology) February 26, 2021 / Un an de confinement : quelques dates qui ont marqué la vie des Belges (chronologie) 26 février 2021 [https://www.lespecialiste.be/fr/actualites/un-an-de-confinement-quelques-dates-qui-ont-marque-la-vie-des-belges-chronologie.html?fbclid=IwAR1zxNCGrZteJ29ETcy7Jkk1UKe0oJsQLHrANIbTKBl6IZOYphlk8R6U2zc]. Accessed 14 July 2024. Data engineering in GIS: let the ETL journey begin-ETL - Extract, Transform, Load / Ingénierie des données dans les SIG: que le voyage ETL commence- ETL – Extract, Transform, Load. [https://gisgeography.com/data-engineering-gis/]. Accessed 14 July 2024. Github: Package Cartography. [https://github.com/riatelab/cartography]. Accessed 14 July 2024. Zirar W, ticsante-com.: Hospital health data: the Healthcare Data Institute calls for the creation of a "national observatory" / Données de santé hospitalières: le Healthcare Data Institute appelle à la création d'un "observatoire national" . Ticsantecom 2023, 81 .[https://www.ticsante.com/story?ID=6696]. Accessed 14 July 2024 The medical photo library: the first solution for managing medical photos. Pixacare automates filing and secures photos / La photothèque médicale: première solution de gestion des photos médicales. Pixacare automatise le classement et sécurise les photos. [https://en.pixacare.com/a-propos]. Accessed 14 July 2024. Additional Declarations No competing interests reported. Supplementary Files SF1supplementaryfile1.zip Appendix.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4796017","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":336992548,"identity":"7d92d5c5-df1b-4273-bbf7-2401b45c4cee","order_by":0,"name":"Anne QUESNEL-BARBET","email":"data:image/png;base64,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","orcid":"","institution":"1-Lille University Hospital (CHU de Lille), Regional House of Clinical Research (MRRC), Public Health.","correspondingAuthor":true,"prefix":"","firstName":"Anne","middleName":"","lastName":"QUESNEL-BARBET","suffix":""},{"id":336992549,"identity":"ef87f1c2-0b98-4653-ac7d-59c0e4e97834","order_by":1,"name":"Thierry PAGES","email":"","orcid":"","institution":"2-TITSOFT","correspondingAuthor":false,"prefix":"","firstName":"Thierry","middleName":"","lastName":"PAGES","suffix":""},{"id":336992550,"identity":"1ddfa383-aa8d-4a8c-a364-ae8179968dd3","order_by":2,"name":"Julien SOULA","email":"","orcid":"","institution":"3-CERIM Lab - EA 2694: Public Health, Faculty of Medicine, University of Lille.","correspondingAuthor":false,"prefix":"","firstName":"Julien","middleName":"","lastName":"SOULA","suffix":""},{"id":336992551,"identity":"11ea7e3c-0186-42b1-9526-d05623fd2fd4","order_by":3,"name":"Gilles MAIGNANT","email":"","orcid":"","institution":"4-RETINES Laboratory - Faculty of Medicine, University of Côte d'Azur (UCA).","correspondingAuthor":false,"prefix":"","firstName":"Gilles","middleName":"","lastName":"MAIGNANT","suffix":""},{"id":336992552,"identity":"57db09df-4771-4074-ae52-96e889153d24","order_by":4,"name":"Arnaud HANSSKE","email":"","orcid":"","institution":"5-DSI/CIO/CMIO - Catholic Hospitals of Lille (GHICL), Nord.","correspondingAuthor":false,"prefix":"","firstName":"Arnaud","middleName":"","lastName":"HANSSKE","suffix":""}],"badges":[],"createdAt":"2024-07-24 14:10:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4796017/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4796017/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":62104064,"identity":"c5e6a45f-872e-4ea7-a2cf-ea6014573b8e","added_by":"auto","created_at":"2024-08-09 10:18:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":213595,"visible":true,"origin":"","legend":"\u003cp\u003eSynthetic (a) and detailed (b) webmapping architectures of uMap atlases for France\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/420401978d838341da73d2b9.png"},{"id":62104062,"identity":"2daf5474-2b1e-403a-ba0d-3df8bcbfca6f","added_by":"auto","created_at":"2024-08-09 10:18:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":116143,"visible":true,"origin":"","legend":"\u003cp\u003eData transformation diagram\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/ea8d2b4f7beca11758aaa983.png"},{"id":62103049,"identity":"b1219187-93a9-4503-8063-ba7764702f3a","added_by":"auto","created_at":"2024-08-09 10:10:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":122898,"visible":true,"origin":"","legend":"\u003cp\u003eData pipeline, geomatic processing software system - step 1 - without webmapping\u003c/p\u003e\n\u003cp\u003eLegend of Figure 3 a distinct shape for each type of information: rectangle: input-output-result; rhombus or diamond: software-processing; ellipse: QGis-workflow-modeler.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/ce0da5b3aa8321e53600e985.png"},{"id":62103052,"identity":"6e0be5b7-fdb4-46fe-b89d-92d31049a24f","added_by":"auto","created_at":"2024-08-09 10:10:50","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":138104,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow QGis (\u003cem\u003e.model3\u003c/em\u003e) “part1 - obtaining the TrafficLight\u003cem\u003e.geojson\u003c/em\u003emap for uMap “, step 1-without-webmapping\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/cc07ed34bf7800f3b5943df4.png"},{"id":62103053,"identity":"af465583-7591-46e4-88dd-707e1b650a37","added_by":"auto","created_at":"2024-08-09 10:10:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":36288,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow QGis (\u003cem\u003e.model3\u003c/em\u003e) “part 2 - batch processing - TrafficLight.\u003cem\u003egeojson\u003c/em\u003efor uMap” step 1-without-webmapping\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/5c01dc55bcdd6575f69e08dd.png"},{"id":62103060,"identity":"a2efb16c-c76f-482f-9e93-0f5ba80d8599","added_by":"auto","created_at":"2024-08-09 10:10:52","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":60807,"visible":true,"origin":"","legend":"\u003cp\u003eData pipeline, geomatic processing software system – step 2-with-webmapping.\u003c/p\u003e\n\u003cp\u003eFigure 6 legend: A distinct shape is associated with each of the two types of information: diamond: processing; rectangle:\u003cstrong\u003e \u003c/strong\u003einput-output-result.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/75187bcdc3ecb152ce2986eb.png"},{"id":62103056,"identity":"ddb9142b-b6a6-4e67-a672-593aaa5f9b75","added_by":"auto","created_at":"2024-08-09 10:10:51","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":229399,"visible":true,"origin":"","legend":"\u003cp\u003eDouble caption data pipeline - France and region, step 2-with-uMap-webmapping\u003c/p\u003e\n\u003cp\u003eLegend of Figure 7: Three blocks of procedures steps 1 to 8. A distinct shape is associated with three types of information: \"rectangle: input-output; cylinder: database and diamond: software-processing\".\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/ef72817609d8e20b05b536da.png"},{"id":62104065,"identity":"6810c9f1-8515-456e-8bef-e042734b2ad4","added_by":"auto","created_at":"2024-08-09 10:18:51","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":129726,"visible":true,"origin":"","legend":"\u003cp\u003e1\u003csup\u003est\u003c/sup\u003e workflow -Win cmd/XnViewMP, 2\u003csup\u003end \u003c/sup\u003e-XnConvert_thumbnail processing for Photoshop 7 – step 2-with-webmapping\u003c/p\u003e\n\u003cp\u003eLegend Figure 8: Two blocks of procedures steps 1 to 3. A distinct shape is associated with three types of information: \"rectangle: input-output; ellipse: workflow process and diamond: software-processing\".\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/f9825b8b7109e1fa9b1bdeda.png"},{"id":62103062,"identity":"b67c29a8-797d-413d-9cdb-3c3dc5a045fb","added_by":"auto","created_at":"2024-08-09 10:10:52","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":200677,"visible":true,"origin":"","legend":"\u003cp\u003ePhotoshop7 workflow for legends with and without QR Code, step 2-with-webmapping\u003c/p\u003e\n\u003cp\u003eLegend Figure 9: One block of procedures. A distinct shape is associated with three types of information: \"rectangle: input-output; ellipse: workflow process and diamond: software-processing\".\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/9a361a4241c8a4415cb29299.png"},{"id":62103057,"identity":"94d8e879-ea43-4d99-b0f5-ffeed40c7091","added_by":"auto","created_at":"2024-08-09 10:10:51","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":409050,"visible":true,"origin":"","legend":"\u003cp\u003eExcess mortality map by TrafficLight (map of atlas) output.kml format - realized with Philcarto\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/ad72c96e73114b1f361fa537.png"},{"id":62103061,"identity":"618e1f05-abf5-4c8e-9df1-baea78aa349d","added_by":"auto","created_at":"2024-08-09 10:10:52","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":587493,"visible":true,"origin":"","legend":"\u003cp\u003euMap atlas by TrafficLight, France, \u003ca href=\"http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.572/4.944\"\u003e\u003cem\u003eaddress URL the atlas\u003c/em\u003e.\u003c/a\u003e\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/4faef9116c1dc313604059b8.png"},{"id":62103058,"identity":"c3fb9cf1-3f4a-4c73-afb2-e12401274946","added_by":"auto","created_at":"2024-08-09 10:10:51","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":390299,"visible":true,"origin":"","legend":"\u003cp\u003eTrafficLight atlas, zoom on the south-west-regions - Bayonne town tooltip.\u003cem\u003e \u003c/em\u003e\u003ca href=\"http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.572/4.944\"\u003e\u003cem\u003eURL address of the atlas.\u003c/em\u003e\u003c/a\u003e\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/468f8fb2114cfcb9172e0729.png"},{"id":62105392,"identity":"60300355-7376-4a4a-834d-b8250bdfc87c","added_by":"auto","created_at":"2024-08-09 10:34:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8037791,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/5ea5ef2f-66cf-477b-b470-f49297b7dc2e.pdf"},{"id":62103059,"identity":"f497c1c5-2956-47cc-8470-ad37d760db43","added_by":"auto","created_at":"2024-08-09 10:10:52","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3715076,"visible":true,"origin":"","legend":"","description":"","filename":"SF1supplementaryfile1.zip","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/e4f2e93944618ccbcf2dfc50.zip"},{"id":62104851,"identity":"30b6dbea-158c-4a1f-9e51-17d363ced0cc","added_by":"auto","created_at":"2024-08-09 10:26:50","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":18752,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-4796017/v1/bc46d2f6265cf1b100adcd68.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Relevance Of The uMap Collaborative Platform As A Support For Choropleth Mapping: An Atlas Applied To All-Cause Excess Mortality Alerts By Traffic Light - 1st French Containment In 2020","fulltext":[{"header":"1. Introduction ","content":"\u003cp\u003eThe needs and interests in geomatics of the medical sector, particularly hospitals, have led us over time to strengthen our interdisciplinary collaborations on joint research projects. In response to the health crisis, our new health geomatics project led us to create a mortality atlas targeting \u0026quot;all pathologies that can be visualized on very fine grids of the communes and arrondissements of Marseille, Lyon and Paris on the scale of France\u0026quot;.\u003c/p\u003e\n\u003cp\u003eOur monitoring of geomatics technologies, IT development, and data sources has led us to focus on two atlas environments.\u003c/p\u003e\n\u003cp\u003eOur main three-point objective is firstly to provide an original and relevant decision-making aid in a geomatics environment (on a French scale), in the form of a cartographic atlas whose information is processed at a fine territorial grid (communes and arrondissements). This processed information then provides new insights compared with studies whose information is aggregated at coarser departmental and regional levels, leading to visualization biases e.g. a mesh that is too aggregated can hide, render invisible or imperceptible other information remaining to be discovered [1].\u003c/p\u003e\n\u003cp\u003eThe second point is to process information on mortality in 2020 (compared with 2018 and 2019) based on the period of the first French containment.\u003c/p\u003e\n\u003cp\u003eFinally, the third point is to make the thematic information mapped on the scale of France and its (\u003cem\u003eD\u0026eacute;partements et R\u0026eacute;gions d\u0026apos;Outre-Mer\u0026nbsp;\u003c/em\u003e(DROMs/OSDRs, the OverSeas Departments and Regions)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eeasily understandable to the general public and users of our atlases in the uMap environment\u003cem\u003e.\u0026nbsp;\u003c/em\u003eTo this end, we have constructed atlas-based statistical indicators of excess mortality. These over-under mortality indicators (per 10,000 inhabitants at municipal level) are also displayed in the tooltips.\u003c/p\u003e\n\u003cp\u003eUsers can connect to the uMap environment for each atlas instance associated with a statistical variable, in four layers representing four major regions: northeast, northwest, southeast and southwest.\u003c/p\u003e\n\u003cp\u003eThe main methodological approaches for obtaining results are divided into two stages: \u0026quot;step 1, geomatics without webmapping\u0026quot; and \u0026quot;step 2, geomatics with webmapping\u0026quot; [2].\u003c/p\u003e\n\u003cp\u003eOur atlas instances (uMap showcases) are the main results of our study. Among the three e-atlases, the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) thematic atlas created from the \u0026ldquo;Khi2LnFr\u0026rdquo; variable, displays one of three colors: red, orange, and green, if the commune is exclusively concerned by an excess mortality alert. The latter atlas appears even more relevant, as cartographic information is instantly visible and easily interpreted by the human eye. We have qualified this \u003cem\u003e\u0026quot;FeuTricolore\u0026quot;\u003c/em\u003e (TrafficLight) atlas as the best messenger.\u003c/p\u003e\n\u003cp\u003eAfter a review of the existing geomatics environment (history of digital cartography, its standards and health geomatics), this article describes the objective reasons for using the uMap geomatics environment as a support for thematic health atlases (composed of choropleth maps). It describes the methods and processes involved in developing the uMap environment, followed by results on uMap usability (platform ergonomics) and cartography (information visualization and interpretation). Now, we present discussions on the geomatic process (uMap software package), and on the advantages and disadvantages in terms of design and use. The outlook then looks at how the uMap environment and the tasks can be assessed in terms of its performance, attractiveness, and profitability, and what new developments, geomatic instantiations and other improvements to the environment can be envisaged. We end with our conclusions.\u003c/p\u003e"},{"header":"2. Background","content":"\u003ch2\u003e2.1. State of the art in geomatics, healthcare geomatics\u003c/h2\u003e\n\u003ch3\u003e2.1.1. History of digital cartography \u003c/h3\u003e\n\u003cp\u003eGeography is an integral part of our modern digital world, and the map is its main tool. The map is a model of partial representation of the geographic space under analysis; many different types of maps are produced, with varying degrees of respect for scientific reasoning and graphic semiology standards [3-5]. Nowadays, maps are most often produced using \u003cem\u003eCartographie et Dessin Assistés par Ordinateur\u003c/em\u003e (\u003cem\u003eCAO-DAO/CAMD \u003c/em\u003eComputer Aided Mapping and Drawing) software, or from software environments with multiple functionalities for spatial representation and analysis, such as \u003cem\u003eSystèmes d’Information Géographique \u003c/em\u003e(\u003cem\u003eSIGs/GISs,\u003c/em\u003e Geographic Information Systems) and shared geographic information platforms: all these environments represent the world of geomatics.\u003c/p\u003e\n\u003cp\u003eGeomatics is a contraction of the words \"geography\" and \"computer science» and is associated with the world of digital cartography and spatial analysis. According to Henri Pornon, it is the discipline that deals with the management of spatially referenced data, involving the sciences and technologies associated with their acquisition, storage, processing, and dissemination. A distinction is made between \"geographic data\", representative of an object in the territory, and \"localized data\", representative of an object positioned in the territory [6]. By contraction of geographic and localized, we speak of geolocalized data for positioned object.\u003c/p\u003e\n\u003cp\u003eIn other words, geomatics is a group of IT methods and tools designed to collect, organize, analyze, and represent - in the sense of modeling - geolocalized data. It enables geolocalized information to beacquired, structured, integrated and analyzed, before the results are published in the form of maps. Geomatics is generally associated with \u003cem\u003eSIG/GIS \u003c/em\u003eas the main digital tool implementing the concepts and methods of geography [6].\u003c/p\u003e\n\u003cp\u003eGeomatics has no existence without geolocalized data, which has been evolving since the 1970s, with the production, distribution, and use of public geolocalized data[6]. The recent concept of \u003cem\u003eTrouvable, Accessible, Interopérable, Réutilisable\u003c/em\u003e \u003cem\u003edonnées \u003c/em\u003e(\u003cem\u003eFAIR,\u003c/em\u003e Findable, Accessible, Interoperable, Reusable data) [7, 8] is defined as the ability for IT systems (with little or no human intervention) to manage information, in compliance with the four rules: Findability, Accessibility, Interoperability and Reusability. The quality of data production is also under scientific scrutiny in the collaborative \u003cem\u003eProjet OpenStreetMap, carte du monde modifiable gratuitement\u003c/em\u003e (\u003cem\u003eOSM,\u003c/em\u003e OpenStreetMap project: a free editable map of the world) [9-11].\u003c/p\u003e\n\u003cp\u003eNumerous concepts have also emerged from this ongoing development of rigorous data management, that impacts geolocalized data by inheritance. We refer to open data - free databases[1] [12, 13], crowdsourcing and crowd mapping - participative production and mapping [14-16]; to open-source - free and open computer code with authorized modification and redistribution; to open source model - provision of software and open/free code as opposed to proprietary code [17, 18]; or to participatory Web 2.0 or contributory internet - shared information and geographic information platform; collaborative platform; information commons and commons [19].\u003c/p\u003e\n\u003cp\u003eLocation-based information and free geomatics[2] [9] are by heritage affected by the \"4 rules of \u003cem\u003eFAIR\u003c/em\u003e data, Findable, Accessible, Interoperable, Reusable\" of free computing [7, 8]. These include interoperability, which guarantees (thanks to open standards[3]) the exchange of files between users equipped with different hardware or software [20]. Two types of interoperability can be distinguished: syntactic interoperability, allowing two systems to communicate with one another, and inter-domain interoperability, allowing several organizations to cooperate and exchange information with one another [21]. Interoperability thus offers proprietary producers the opportunity to turn to these concepts, gradually leading to the emergence of production chains and pipelines[4] that are completely free to use [9, 22, 23].\u003c/p\u003e\n\u003cp\u003eThree categories of open standards accompany interoperability[24]. \u003cstrong\u003e(1)- \u003c/strong\u003estandards set by software publishers and users (e.g., the \"shapefile \u003cem\u003e.shp\u003c/em\u003e\" file format from ©ESRI, a \u003cem\u003eSIG/GIS\u003c/em\u003e publisher, or the \"DraWinG \u003cem\u003e.dwg\u003c/em\u003e\u003cstrong\u003e\" \u003c/strong\u003efile format from ©AutoCAD, a drawing software publisher). \u003cstrong\u003e(2)- \u003c/strong\u003eformal standardsdefined according to a strict protocol by standards bodies such as \u003cem\u003ethe Consortium Géospatial Ouvert \u003c/em\u003e(\u003cem\u003eOGC\u003c/em\u003e, Open Geospatial Consortium (\u003cem\u003eOGC\u003c/em\u003e[5], 1994)) [7] which represents a worldwide resource community for geospatial information and standards. Other representative bodies include the \u003cem\u003eAssociation Française de NORmalisation \u003c/em\u003e(\u003cem\u003eAFNOR/FSA\u003c/em\u003e, French Standards Association) [25], \u003cem\u003eOrganisation Internationale de Normalisation (ISO\u003c/em\u003e, International Organization for Standardization) [26], the \u003cem\u003eComité Européen de Normalisation\u003c/em\u003e (\u003cem\u003eCEN/ECS\u003c/em\u003e, European Committee for Standardization) and Etalab [27] of \u003cem\u003ethe Département de la Direction Interministérielle du NUMérique \u003c/em\u003e(\u003cem\u003eDINUM/IDD\u003c/em\u003e, Interministerial Digital Department), whose aim is to improve public service and public action through data; \u003cstrong\u003e(3)- \u003c/strong\u003eFrench government standards created to adapt uses to the French national territory. Thus, the \u003cem\u003eConseil National de l'Information Géolocalisée \u003c/em\u003e(\u003cem\u003eCNIG/NCGI\u003c/em\u003e, National Council for Geolocated Information) - newly renamed (\u003cem\u003eCNIG for geographic \u003c/em\u003ebecomes \u003cem\u003eCNIG for Geolocalized/Geolocated\u003c/em\u003e) to cover the topography of territories and, more recently, their location - coordinates and supports public and private players in meeting the evolving challenges of standardization, innovation, production, distribution and sharing of geolocated information. Finally, standards bodies such as the\u003cem\u003e Consortium Géospatial Ouvert \u003c/em\u003e(\u003cem\u003eOGC,\u003c/em\u003e Open Geospatial Consortium) and the \u003cem\u003eConseil National de l'Information \u003c/em\u003eGéolocalisée (\u003cem\u003eCNIG/NCGI,\u003c/em\u003e National Council for Geolocated \u003cem\u003eInformation\u003c/em\u003e) comply with European policies, directives and regulations: \u003cem\u003eInfrastructure d’Information Spatiale en Europe \u003c/em\u003e(\u003cem\u003eINSPIRE Directive\u003c/em\u003e, Infrastructure for Spatial Information in Europe), the European fair and innovative data (DataAct) [28].\u003c/p\u003e\n\u003cp\u003eThe following is a selection of major projects and standards impacting geomatics:\u003c/p\u003e\n\u003cp\u003e1. The standards of the international association of \u003cem\u003eProducteurs de Pétrole et de Gaz \u003c/em\u003e(\u003cem\u003ePPG/OGP\u003c/em\u003e, Oil and Gas Producers), formerly\u003cem\u003e Groupe Européen d’Etude du Pétrole \u003c/em\u003e(\u003cem\u003eEPSG\u003c/em\u003e, European Petroleum Survey Group) 1985 for geodesy maintenance and free sharing of the \u003cem\u003eEPSG\u003c/em\u003e geodetic parameter dataset [29-32].\u003c/p\u003e\n\u003cp\u003e2. Web standards[6], including the \u003cem\u003eWorld Wide Web Consortium \u003c/em\u003e(\u003cem\u003eW3C\u003c/em\u003e) for developers and by heritage for geomatics developers [13, 24, 33].\u003c/p\u003e\n\u003cp\u003e3. National projects based on very large-scale building information modeling standards (BIM and GeoBIM) [34, 35].\u003c/p\u003e\n\u003cp\u003e4. \u003cem\u003eThe Groupe de Travail sur l’Information Géospatiale de la Défense \u003c/em\u003e(\u003cem\u003eGTIGD/DGIWG,\u003c/em\u003e The Defence Geospatial Information Working Group) has been managing geospatial information and ensuring interoperability since 1983 by creating the standards, implementation guidelines and procedures necessary to enable the provision, exchange and use of standardized geospatial information [36].\u003c/p\u003e\n\u003ch3\u003e2.1.2. Geomatics in healthcare - themes, visualization, territorial networking\u003c/h3\u003e\n\u003cp\u003eAs seen above, geomatics is benefiting from a continuous digital transformation. The map is at the center of attention in many fields of business and research. Open-source geomatics is accelerating this trend, meaning that cartography is no longer the sole preserve of geolocation information specialists. Many professionals, such as IT specialists, data scientists, open-source project contributors, etc., are now involved in geomatics [9] and scientists (whether or not they specialize in location-based information) have seized on the map as a pretty, useful, strategic, and valuable object; an object supporting knowledge, spatial memorization and reflection and/or as a digital object to be produced, automated and distributed.\u003c/p\u003e\n\u003cp\u003eGeomatics in healthcare has become increasingly important in recent decades. Current societal developments, marked by the covid 19 pandemic, have boosted the use of digital maps in healthcare. At the height of the crisis, maps were followed by millions of Internet users and other readers. This societal awareness of health risks, accentuated by the covid pandemic, calls forthe map to become an essential part of our societies and Public Health.\u003c/p\u003e\n\u003cp\u003eNew map showcases - based on geomatics technologies - are emerging in the form of online dashboards, replacing more complex \u003cem\u003eSIG/GIS \u003c/em\u003eenvironments. These dashboards can be public, private or mixed (the private part is accessible with authorization), and offer users visualization and interaction functionalities [37, 38]. Static or dynamic atlases are integrated into these environments, with geographical scales ranging from global to local, using geolocalized health data that are more or less aggregated according to the authorized territorial mesh for spatio-statistical analysis, which can lead to a number of biases [1].\u003c/p\u003e\n\u003ch4\u003e2.1.2.1. Cartographic information bias - four causes of over-aggregated meshes\u003c/h4\u003e\n\u003cp\u003eThe use of meshes that are too aggregated in cartography can lead to information bias for four reasons identified by the authors [39]. \u003cstrong\u003e(1) \u003c/strong\u003eA loss of accuracy can occur when using a mesh that is too large. \u003cstrong\u003e(2) \u003c/strong\u003eThe \"aggregation problem\" arises if the analysis is misinterpreted due to data appearing more homogeneous than they actually are. \u003cstrong\u003e(3) \u003c/strong\u003eThere may be difficulties in visualizing inter-zone disparities or geographical variations. \u003cstrong\u003e(4) \u003c/strong\u003eFinally, a phenomenon of exaggeration or over-representation of information for sparsely populated areas may appear due to an overly wide mesh wrongly accentuating the sanitary phenomenon [40].\u003c/p\u003e\n\u003cp\u003eAs a result, maps with mesh sizes that are too aggregated can lead to errors of interpretation and planning, and thus to ineffective public health interventions, for example. It is therefore important to use maps with the most appropriate mesh size possible, in order to address the limitations outlined above [41].\u003c/p\u003e\n\u003ch2\u003e2.2. Interests and needs identified by the hospital sector for geomatics in healthcare \u003c/h2\u003e\n\u003cp\u003eIn the literature, scientific publications in the healthcare sector bear witness to the growing interest in geomatics; this interest, accentuated by the covid health crisis, was already focused before the pandemic on various medical and surgical disciplines, or on public health themes such as assistance with prevention and healthcare organization [6, 42-44].\u003c/p\u003e\n\u003ch3\u003e2.2.1. Point of view the hospital group of the Institut Catholique de Lille\u003c/h3\u003e\n\u003cp\u003eGeomatics offers numerous advantages for representing health indicators in the hospital sector, particularly in evaluating epidemiological surveillance and assessing patient retention or outflow, alongside recent developments in the medico-economic aspects of care pathways. Geomatics indeed helps understanding morbidity and mortality indicators and addressing public health needs within a specific geographic area based on population responsibility.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEpidemiological monitoring offers several advantages for the \u003cem\u003eGroupement des Hôpitaux de l'Institut Catholique de Lille (GHICL\u003c/em\u003e,Group of hospitals of the Institut Catholique de Lille)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e· It enables better adjustment of care production strategies and improves capacity planning for patient care, concretely demonstrated by the creation of bed-manager positions. This allows the healthcare establishment to optimize resource allocation for various treatments—medical, surgical, or obstetrical—by adjusting the number of hospital beds and the necessary medical and nursing staff. It enhances logistical management of home care, including the optimization of scheduling for healthcare professionals attending to patients receiving home hospitalization. \u003c/p\u003e\n\u003cp\u003e· Additionally, geomatics can be utilized to assess the quality of care by analyzing geospatial data. This is crucial for monitoring epidemics and overseeing patient care, whether by the originating facility or other nearby healthcare structures.\u003c/p\u003e\n\u003ch3\u003e2.2.2. State of the art in mortality atlases\u003c/h3\u003e\n\u003cp\u003eGenerally speaking, epidemiological statistics and the mortality atlases they produce are essential tools for understanding mortality patterns and trends in a given region, zone or country and for comparing the mortality rates of the studied areas [45]. \u003c/p\u003e\n\u003cp\u003eIn France,for example, the \u003cem\u003eInstitut National de la Statistique et des Études Économiques (INSEE\u003c/em\u003e\u003cem\u003e, \u003c/em\u003ethe French National Institute of Statistics and Economic Studies), systematically publishes an interpretation of the country's mortality data, as well as several mortality atlases at sub-national (regional or departmental) level. Overall, these atlases show mortality rates by cause of death, age and sex. They provide a better understanding of geographical variations and identify the most vulnerable population groups [46]. Two French institutions stand out: \u003cem\u003eSanté Publique France (SPF/FPH \u003c/em\u003eFrance Public Health) (formerly \u003cem\u003eINstitut de Veille Sanitaire-INVS\u003c/em\u003e) and its “Géodes” platform [47]. \u003cem\u003eSPF/FPH \u003c/em\u003eis an organization in charge of continuous monitoring of the population's state of health and its evolution [48]; the “Sentinelles” network, which monitors the health of nineteen pathologies (infectious, seasonal, etc.) [49]. In June 2020, the France 3 media highlighted the fact that \u003cem\u003eneither the SPF/FPH \u003c/em\u003enor \u003cem\u003ethe Agence Régionale de Santé \u003c/em\u003e(\u003cem\u003eARS/RHA\u003c/em\u003e\u003cstrong\u003e, \u003c/strong\u003eRegional Health Agency) publishes epidemiological data on a commune-by-commune basis. Daily reports are limited to deaths occurring in hospitals, as well as in the \u003cem\u003eEtablissements d'Hébergement pour Personnes Âgées Dépendantes (EHPAD/REFDE, \u003c/em\u003eResidential Establishments For The Dependent Elderly), for which deaths are aggregated and transmitted only at regional level [50].\u003c/p\u003e\n\u003cp\u003eAt global level,\u003cem\u003ethe Organisation Mondiale de la Santé (OMS/WHO, World \u003c/em\u003eHealthOrganization) also produces mortality atlases for specific countries and regions. These mortality atlases illustrate and present data on causes of death, mortality rates, as well as temporal trends and the development of these axes in different countries and regions of the world. This enables the \u003cem\u003eOMS/WHO \u003c/em\u003eto monitor global mortality, assess progress in public health, as well as develop strategies to reducing these mortality rates [51].\u003c/p\u003e\n\u003cp\u003eThese national and internationalatlases are important tools for decision-makers, be they politicians or specialists involved in health, researchers or professionals at all levels working to improve public health. \u003c/p\u003e\n\u003ch2\u003e2.3. The POLESAT health geomatics project and its latest tool, the uMap mortality atlases\u003c/h2\u003e\n\u003cp\u003ePOLESAT stands for \"\u003cem\u003ePôles Sanitaires\u003c/em\u003e\" with the capitalized acronym designating our overall project, an innovative e-geo-platform that will evolve over time, aimed at both the general public and professionals [52, 53]. POLESAT currently comprises four modules: an e-atlas dedicated to the visualization of hospital care supply and demand [54] a medicalized geographical support dedicated to patient orientation and programmed medical choice [55] a prospective health planning tool with variable scenario simulation geometry, available in two versions: \"e-PoleSat-démo\", a public version, and \"e-PoleSat-métier\", a private version [52, 53]. The all-cause mortality uMap atlas environment is the fifth POLESAT module.\u003c/p\u003e\n\u003ch3\u003e2.3.1. The new mortality atlas environments of the overall POLESAT project\u003c/h3\u003e\n\u003cp\u003eAll-cause mortality atlases in the uMap environment are hosted on the umap.openstreetmap.fr server [56, 57, 58] and for the \u003cem\u003eGroupement des Hôpitaux de l'Institut Catholique de Lille (GHICL\u003c/em\u003e,Group of hospitals of the Institut Catholique de Lille) on the \u003cem\u003eghicl\u003c/em\u003e.net server. These environments have been invested over time by the team of developers and geomaticians.\u003c/p\u003e\n\u003cp\u003eWorking in two environments, requiring different geomatics and development skills, enabled us to compare and reflect on the technical approaches used and the graphic and cartographic renderings obtained. Comparisons focused on database processing, development and geomatics tasks with remote server management, and geomatics tasks without remote server management [59]. These comparisons concerned two types of \"workflows\"[7], one based on a \"software chain\" and the other on \"a single software environment\":\u003c/p\u003e\n\u003cp\u003e· The first workflow produces two types of data pipelines [22 , 23] \u003cem\u003eCartographie et Dessin Assistés par Ordinateur \u003c/em\u003e(\u003cem\u003eCAO-DAO/CAMD,\u003c/em\u003e Computer Aided Mapping and Drawing) and \u003cem\u003eSIG/GIS\u003c/em\u003e including a source code editor; an \u003cem\u003eEnvironnement de Développement Intégré (IDE\u003c/em\u003e,integrated development environment) for R [60] (a programming language for statistical calculations and graphics), spreadsheets etc. ;\u003c/p\u003e\n\u003cp\u003e· The second workflow is internal to the \u003cem\u003eSIG/GIS \u003c/em\u003eQGis software environment, integrating specialized task automation tools [61].\u003c/p\u003e\n\u003cp\u003eFinally, the literature lists some very positive comparisons and reflections on uMap such as \u003cstrong\u003e(1) \u003c/strong\u003edata management in the context of collaborative mapping; \u003cstrong\u003e(2) \u003c/strong\u003edata confidentiality in the context of activist mapping; \u003cstrong\u003e(3) \u003c/strong\u003econtributor management in the context of \"crowd mapping\" [16]; and \u003cstrong\u003e(4) \u003c/strong\u003eopen-source philosophy in the context of teaching collaborative mapping online [59 , 62].\u003c/p\u003e\n\u003cp\u003eThis led us to an overall assessment of the advantages and disadvantages of the two working environments, and to ideas for future versions. uMap is the only environment presented in the article.\u003c/p\u003e\n\u003ch2\u003e2.4. Reorientation of the original idea of an atlas to combat the covid-19 pandemic\u003c/h2\u003e\n\u003cp\u003eThe atlas of all-cause mortality at regional and French national levels began during the period of the first French containment of the covid 19 health crisis. Our primary objective was to contribute, like so many others and within our means, to the fight against the pandemic. To do this, we wanted to show customers, patients, and hospital professionals the lethal epidemic impact and excess mortality of covid-19 at a fine territorial mesh, in real time. However, after monitoring the situation (both scientifically and in the media), and re-evaluating our resources (both human and technical), we reoriented our primary objective towards:\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e1- \u003c/strong\u003eThe use of all-cause disease mortality data [63] which offers a number of advantages: the data are finely disseminated at the local level; access is open (open-data); they make it possible to bypass young covid databases with non-negligible biases, such as \"errors in coding deaths by covid-19 diagnosis\" [64]\",\u003c/p\u003e\n\u003cp\u003e2- A geomatic rather than an epidemiological approach: since we wanted to test geomatic technologies for relevant visualization of statistical mortality indicators and dissemination of geolocalized information.\u003c/p\u003e\n\u003cp\u003eIn addition, the scientific and media monitoring carried out as we progressed through the covid-19 crisis in 2020 showed that:\u003c/p\u003e\n\u003cp\u003e1- French epidemiological data on the incidence of covid-19 became unavailable at the commune level, and the few covid cluster maps available at the commune level eventually disappeared from the French media. Belgium disseminated covid incidence per 100,000 inhabitants at the commune level [65],\u003c/p\u003e\n\u003cp\u003e2- Many French studies and their associated maps were based on highly aggregated covid-19 open data [66].\u003c/p\u003e\n\u003cp\u003e[1] The Open-Source world is characterized by the freedom to use, copy, study, modify and redistribute. The challenges of free software are 1-liberty, 2-perpetuity, 3-interoperability, 4-quality-reliability-security.\u003c/p\u003e\n\u003cp\u003e[2] Free geographic information is a formalized product produced within a non-institutional framework, in a specific ecosystem, according to the tried and tested methods of scientific work, namely the rectification process, the reproducibility of results and the verifiability of information.\u003c/p\u003e\n\u003cp\u003e[3] An open standard is any interoperable communication, interconnection or exchange protocol and data format whose technical specifications are publicly available and unrestricted in terms of access and implementation.\u003c/p\u003e\n\u003cp\u003e[4] Here's a definition of a data pipeline: A data pipeline encompasses a series of actions that begin with the ingestion of all raw data from any source, to rapidly transform it into data ready for exploitation.\u003c/p\u003e\n\u003cp\u003e[5] \"OGC standards are the cement of geospatial information interoperability, used by thousands of organizations worldwide and represented in millions of lines of code\".\u003c/p\u003e\n\u003cp\u003e[6] As Roger Johansson said in 2004: \"Web standards are technologies established by the \u003cem\u003eW3C\u003c/em\u003e and other standards bodies for creating and interpreting Web content. These technologies are designed to create documents that are durable and accessible to all.\u003c/p\u003e\n\u003cp\u003e[7] \"Workflow\" is an earlier IT concept, closely related to \"pipeline\". Workflow is more related to human activity than to machine activity. However, processing chains within software environments are called workflows, as in QGis for example.\u003c/p\u003e"},{"header":"3. Material and methods","content":"\u003ch2\u003e3.1. Material\u003c/h2\u003e\n\u003cp\u003eOur data sources, software (open, free, or private), formats and standards are listed below, along with a brief list of software associated with our early project watch.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOpen statistical data from the French National Institute for Statistics and Economic Studies \u003cem\u003e(INSEE)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData from the \u003cem\u003eInstitut National de la Statistique et des \u0026Eacute;tudes \u0026Eacute;conomiques\u0026nbsp;\u003c/em\u003e(\u003cem\u003eINSEE\u003c/em\u003e\u003cem\u003e,\u0026nbsp;\u003c/em\u003ethe French National Institute of Statistics and Economic Studies) are available in spreadsheet format (.\u003cem\u003ecsv\u0026nbsp;\u003c/em\u003eor \u003cem\u003e.xlsx\u003c/em\u003e). Three gross datasets were extracted in May 2020, covering the period from March 1 to May 15, 2020, inclusive (1st French containment of the Covid-19 crisis):\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Deaths by commune representing all-cause mortality of the municipal population (2017 census) for the three years from 2018 to 2020 [63, 67, 68].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The nomenclature of communes, district (\u0026ldquo;arrondissements\u0026rdquo;), departments and regions and the \u003cem\u003eCode Officiel G\u0026eacute;ographique\u0026nbsp;\u003c/em\u003e(\u003cem\u003eCOG/OGC,\u0026nbsp;\u003c/em\u003eOfficial Geographical Code) (2020 vintage) maintained by\u003cem\u003e\u0026nbsp;INSEE\u003c/em\u003e [69].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The 2017 population census data are made up of two files: one in csv format \u0026quot;Communes.csv\u0026quot; listing the population of mainland France and the \u003cem\u003eD\u0026eacute;partements et Territoires d\u0026apos;Outre Mer\u003c/em\u003e \u003cem\u003e(DOM-TOM)\u0026nbsp;\u003c/em\u003erenamed in (\u003cem\u003eD\u0026eacute;partements et R\u0026eacute;gions d\u0026apos;Outre-Mer\u0026nbsp;\u003c/em\u003e(DROMs/OSDRs, the OverSeas Departments and Regions) excluding Mayotte, and one in Excel format \u0026quot;mayotte-RP2017.xls\u0026quot; for Mayotte only [63, 67, 68]. See the deathDataset repository [63, 70] in [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip] and [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/SF1_supplementary_file_1.zip]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOpen geolocalized data from the \u003cem\u003eInstitut National de l\u0026rsquo;Information G\u0026eacute;ographique et Foresti\u0026egrave;re (IGN,\u0026nbsp;\u003c/em\u003eNational Institute of Geographic and Forest Information)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eInstitut National de l\u0026rsquo;Information G\u0026eacute;ographique et Foresti\u0026egrave;re (IGN,\u0026nbsp;\u003c/em\u003eNational Institute of Geographic and Forest Information) raw basemaps for the year 2020, are so-called shape files (Esri\u0026apos;s proprietary \u003cem\u003e.shp\u0026nbsp;\u003c/em\u003efile) which are reworked and adapted for mapping in Philcarto with the help of the \u0026ldquo;Eclat\u0026rdquo; utility\u0026nbsp;[71, 72]. There are two tables of communes in France, one of which is lightened by a less precise mechanism for defining communal boundaries:\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The \u0026quot;admin-express 2020\u0026quot; database at the local level for the whole of France and by territory [69] updated December 17, 2019, files dated March 25, 2020 ADE, license Etalab 2.0.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The referenced set \u0026ldquo;ADE updated to May 18\u003cstrong\u003e\u003cem\u003e,\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e2020\u0026rdquo;, which covers the geographical area of the \u003cem\u003eD\u0026eacute;partements et R\u0026eacute;gions d\u0026apos;Outre-Mer\u0026nbsp;\u003c/em\u003e(DROMs/OSDRs, the OverSeas Departments and Regions\u003cstrong\u003e)\u0026nbsp;\u003c/strong\u003e(still called DOM_TOM in 2020), with the standard (\u003cem\u003eEPSG\u003c/em\u003e_5490) and the universal file format \u003cem\u003e.mid, .mif.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOther open geolocalized data published in \u0026ldquo;data.gouv\u0026rdquo; website\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Geolocalized data layers for uMap with copyright (\u0026copy;OpenStreetMap contributors, ODbL license) [73-77].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData processing software (open: free, open-source, public domain)\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;SQLite version 3.39.4, is a database used to process all data from \u003cem\u003eINSEE\u003c/em\u003e [78, 79].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The R project is a language and software environment dedicated to statistics and graphics [60]. It is multi-platform software, licensed\u003cem\u003e\u0026nbsp;\u003c/em\u003eunder \u003cem\u003ethe licence publique G\u0026eacute;n\u0026eacute;rale (GNU Non-Unix/\u003c/em\u003eGNU\u0026apos;s Not Unix general public license). The R project has a large ecosystem of extensions (packages), an \u003cem\u003eEnvironnement de D\u0026eacute;veloppement Int\u0026eacute;gr\u0026eacute; (IDE\u003c/em\u003e,\u003cem\u003e\u0026nbsp;\u003c/em\u003eintegrated development environment)\u0026nbsp;and a version management system (Git and Svn).\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The RStudio environment includes the R: v4.2.2 program, the libraries (plyr[8]; openxlsx) and RMarkdown, which provides a report of the processing carried out by R (see Figures 1, 2 in section 3.2.) [80, 81, 82 , 83 , 84].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCartographie et Dessin Assist\u0026eacute;s par Ordinateur\u003c/em\u003e (\u003cem\u003eCAO-DAO/CAMD\u0026nbsp;\u003c/em\u003eComputer Aided Mapping and Drawing) software (open or free)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;QGis \u003cem\u003eSIG/GIS\u003c/em\u003e, v3.18.1-Z\u0026uuml;rich\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e[61] is used for our modeling and workflow processing. It integrates code versions: 202f1bf7e5 and the following libraries: Qt: v5.11.2; GDAL: v3.1.4 [85] GEOS: v3.8.1; CAPI v1.13.3 and Proj: Rel. 6.3.2, \u003cem\u003ethe Syst\u0026egrave;me G\u0026eacute;od\u0026eacute;sique Mondial (WGS 84,\u0026nbsp;\u003c/em\u003eWorld Geodetic System 1984); as well as Project-\u003cem\u003eCRS\u003c/em\u003e-ellipsoid_\u003cem\u003eEPSG\u003c/em\u003e_7030;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eSyst\u0026egrave;me de R\u0026eacute;f\u0026eacute;rence des Coordonn\u0026eacute;es (CRS\u003c/em\u003e, Coordinate Reference System) and Project-\u003cem\u003eCRS\u003c/em\u003e \u003cem\u003eEPSG\u003c/em\u003e_4326 May 1st, 2020.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Philcarto v2021.d\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e[86], this mapping software combines a macro language (\u003cem\u003e.pmc\u003c/em\u003e format to automate the opening of work environments) and several export formats \u003cem\u003e(.ai; .emf\u003c/em\u003e), including the main georeferenced \u003cem\u003e.kml\u0026nbsp;\u003c/em\u003eformat.\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026rdquo;Eclat\u0026rdquo; [71, 72] a utility for processing shapefiles in shapefile.\u003cem\u003eshp\u0026nbsp;\u003c/em\u003eformat or universal files in \u003cem\u003e.mid\u003c/em\u003e and \u003cem\u003e.mif\u0026nbsp;\u003c/em\u003eformats.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;MapSharper (JavaScript) for optimizing data in GeoJSON format [87].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Geojson\u003cem\u003e.io\u003c/em\u003e, an online editor for \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003efiles [88].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;XnView and XnConvert for batch image processing, combining a scripting language like \u003cem\u003eLangage de Requ\u0026ecirc;te Structur\u0026eacute;e\u0026nbsp;\u003c/em\u003e\u003cem\u003e(SQL\u003c/em\u003e, structured query language) [89, 90].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;IrfanView for image cropping with batch processing functions\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e[91].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrivate image processing or \u003cem\u003eDessin Assist\u0026eacute; par Ordinateur (DAO/CAD\u003c/em\u003e, Computer Aided Drawing software)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Photoshop\u0026copy; v7.0 (including \u003cem\u003e.atn\u0026nbsp;\u003c/em\u003escripts) for automated legends in uMap\u0026nbsp;[92].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;MapInfo\u0026copy;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eSIG/GIS\u0026nbsp;\u003c/em\u003ev7.0 - old version is used to manage and retrieve basemaps.\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Paint\u0026copy; [93] and Paint\u003cem\u003e.Net\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ecombine the \u003cem\u003eVisual Basic pour Application (VBA,\u003c/em\u003e Visual Basic for Application) language to automate drawing tasks [94, 95].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;SpreadsheetLight\u0026copy; function library used by the \u0026quot;Eclat\u0026quot; utility [96].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFree and open-source collaborative platform\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The uMap interoperable software package (v1.2.1) [97] is a webmapping environment used to represent our France-wide atlases by variable. uMap allows the creation of maps with OpenStreetMap layers\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eIt is a collaborative mapping software platform licensed under the \u003cem\u003eLicence Publique Faites en ce que Vous Voulez (LPFVV/ WTFPL,\u0026nbsp;\u003c/em\u003eDo What The Fuck You Want To Public License) [98] (free software: redistributable and modifiable; \u003cem\u003eGratuit/Libre et Logiciel de Source Ouverte\u0026nbsp;\u003c/em\u003e(GLLSO/FLOSS, Free/Libre and Open-Source Software)\u0026nbsp;[59, 99, 100\u0026nbsp;]. uMap is hosted on the French server openstreetmap.fr\u0026nbsp;[59]\u0026nbsp;with the domain name umap.openstreetmap.fr\u0026nbsp;[57, 101]. It is a Django project maintained by a community of developers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGlobe\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Google Earth Pro\u0026copy; is used to view and export the kml format [102].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode management software (free, GitHub repository)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Visual Studio Code, under Microsoft-MIT-license [96].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;View map extension for VSCode, published by Random Fractals Inc [103].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Git \u003cem\u003eBash\u003c/em\u003e for Windows (script command-line interpreter), provides a Bourne-Again SHell \u003cem\u003e(BASH)\u003c/em\u003e emulation for running Git (a distributed version control system) from the command line [104].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsortium G\u0026eacute;ospatial Ouvert\u003c/em\u003e \u003cem\u003e(OGC,\u003c/em\u003e open geospatial consortium) standards\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Keyhole Markup Language \u003cem\u003e(KML)\u003c/em\u003e,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eGoogle\u0026apos;s proprietary format, now a \u003cem\u003eConsortium G\u0026eacute;ospatial Ouvert (OGC\u003c/em\u003e open geospatial consortium\u003cem\u003e)\u0026nbsp;\u003c/em\u003estandard. The KML specification is associated with only one projection, \u003cem\u003eEPSG\u003c/em\u003e_4326 [105-110].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003cem\u003eGPX\u003c/em\u003e, is a Global Positioning System (\u003cem\u003eGPS)\u003c/em\u003e eXchange Format, to export format from uMap environment and \u003cem\u003eOGC\u003c/em\u003e standard [7, 26].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eInternationale Association des Producteurs de P\u0026eacute;trole et de Gaz\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e(\u003cem\u003ePPG/OGP\u003c/em\u003e, International Oil and Gas Producers Association standards\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;The \u003cem\u003eEPSG\u003c/em\u003e_4326 projection is associated with the KML standard [31].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProprietary, open, standardized and brand-specific formats\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Shapefile (Esri\u0026copy;) [111, 112].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eproprietary, non-standardized, brand-specific formats\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;GeoJSON (RFC 7946 - OpenStreetMap): see the guides and editors of this format [88, 108, 109, 113].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;model3 (modeler\u0026reg; QGis ) [114].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Atn (script-modeling, macro recording from Adobe Photoshop\u0026copy; PS7) [92, 115].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Pcm (Script-modeling, Philcarto macro recording) [86].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComputer-aided drawing \u003cem\u003e(DAO/CAD)\u0026nbsp;\u003c/em\u003esoftware watch (free, open), collaborative platforms -\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ehigh learning curve\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA monitoring period is essential for any project and has led us to partially test the following software\u0026nbsp;[116]:\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;GIMP-V2.1 combines multilingual support (including Python and Scheme) with image manipulation. It is an alternative to proprietary Adobe Illustrator [117].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Inkscape combines a Simple Inkscape Scripting extension and the Python language to automate repetitive drawing tasks [118, 119].\u003c/p\u003e\n\u003cp\u003e\u0026middot; \u0026nbsp; \u0026nbsp; \u0026nbsp;Mapbox\u0026copy; has been tested to geolocalize and convert\u003cem\u003e\u0026nbsp;.svg\u0026nbsp;\u003c/em\u003eto \u003cem\u003e.geojson\u003c/em\u003e data before import into uMap [120].\u003c/p\u003e\n\u003ch2\u003e3.2. Methods\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eWe begin by presenting the software architecture describing the data processing stages (see Figure 1: Synthetic (a) and detailed (b) webmapping architectures of uMap atlases for France), including (Data transformation diagram of Figure 2).\u003c/p\u003e\n\u003cp\u003eIn our study, data management \u003cem\u003e(ing\u0026eacute;nierie des donn\u0026eacute;es)\u003c/em\u003e is an essential process carried out in different software environments, involving data extraction, cleaning and ordering [121], involving specialized concepts and processes such as data pipelines, databases and the \u003cem\u003eExtraire, Transformer et Charger\u003c/em\u003e (\u003cem\u003eETL,\u003c/em\u003e Extract, Transform, Load) process.\u003c/p\u003e\n\u003cp\u003eOur methodological approaches are schematized to enable better monitoring of data pipelines \u0026ldquo;step 1- without webmapping\u0026rdquo; or \u0026ldquo;step 2 - with webmapping\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003e\u0026ldquo;For step 1 - without webmapping\u0026rdquo;, can be seen the data pipeline diagram for transforming .\u003cem\u003ekml\u003c/em\u003e into a .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003efile in Figure 3 and schematics of processing automation in QGis in Figures 4 and 5.\u003c/p\u003e\n\u003cp\u003e\u0026ldquo;For step 2 - with webmapping\u0026rdquo;, can be seen the resizing processing schematic in Mapshaper in Figure 6; the double caption processing diagram in \u003cem\u003eCAO-DAO/CAMD\u003c/em\u003e and uMap in Figure 7; the thumbnail processing diagram in \u003cem\u003eCAO-DAO/CAMD\u0026nbsp;\u003c/em\u003esoftware in Figure 8. Finally, the Photoshop processing automation diagram is shown in Figure 9.\u003c/p\u003e\n\u003cp\u003e3.2.1. Atlas web architecture\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e3.2.2. INSEE death data engineering\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eAll data sources (see section 3.1. Material) share a common identifier: the \u003cem\u003eINSEE\u003c/em\u003e commune code, consisting of five alphanumeric characters \u003cem\u003e(e.g. 2A014)\u003c/em\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eStructured\u003cem\u003e\u0026nbsp;Query Language (SQL\u003c/em\u003e) is used to analyze and process data in SQLite [78, 79]. All processing is automated, with the exception of one that remains in manual mode. The simplified data transformation diagram is shown in Figure 2 below.\u003c/p\u003e\n\u003cp\u003eData in the form of .\u003cstrong\u003e\u003cem\u003ecsv\u0026nbsp;\u003c/em\u003e\u003c/strong\u003efiles are imported into SQLite and then combined to produce an \u003cem\u003e\u0026quot;insee_dc\u0026quot;\u0026nbsp;\u003c/em\u003etable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAn analysis of the fields (or variables) shows that days and months of birth are not systematically filled in. We have calculated the age of each person, with the exception of those for whom the day or month of birth was not provided. The data is then aggregated by \u003cem\u003eINSEE\u003c/em\u003e communal code, calculating the average age (excluding date with birth not provided), the 2020 mortality, the average mortality over the two years 2018 and 2019, and the mortality difference.\u003c/p\u003e\n\u003cp\u003eThe final step in the Figure 2 process is to associate the death data with the reference 2017 municipal population. The database thus created \u0026quot;\u003cem\u003einsee_dc_pop\u003c/em\u003e\u0026quot; is exported in .\u003cem\u003ecsv\u0026nbsp;\u003c/em\u003eformat.\u003c/p\u003e\n\u003cp\u003eTable 1 below shows the reference structure of the data corresponding to the mortality variables.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e Structure of death database (source \u003cem\u003eINSEE\u003c/em\u003e)\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"719\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDescription\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\"\u003e\n \u003cp\u003e\u003cstrong\u003eExample\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eId_com\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eMunicipal code\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026ldquo;01004\u0026rdquo;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eId_dep\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eDepartmental code\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026ldquo;01\u0026rdquo;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eId_reg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eRegional code\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026ldquo;84\u0026rdquo;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eLabel_com\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eMunicipal name\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003eAmb\u0026eacute;rieu-en-Bugey\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eMuni_pop_2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eMunicipal population no. 2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e14,035\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eDc_2018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eNo. of deaths 2018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eDc_2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eNo. of deaths 2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eDc_2018_2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eAverage no. of deaths 2018-2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e39\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eDc_2020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eNo. of deaths 2020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eMortality_diff.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eDifference of deaths (dc_2020 minus dc_2018_2019)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eAverage_age\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eAverage age at death\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"24.930362116991642%\" valign=\"bottom\"\u003e\n \u003cp\u003eSMM_10000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"50%\" valign=\"bottom\"\u003e\n \u003cp\u003eExcess mortality/undermortality per 10,000 inhabitants\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.069637883008358%\" valign=\"bottom\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3\u003e3.2.3.\u0026nbsp;INSEE data engineering in R, Excel for Philcarto\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eSeveral stages of data pre-processing and statistical calculations in R, Excel and Philcarto:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the pre-processing stage\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e(see in 3.2.3., Table 2 -section lines [1] to [21]), the output file \u0026quot;\u003cem\u003einsee_dc_pop.csv\u0026quot;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003efrom Figure 2 requires further pre-processing steps in R and Excel, essentially for the addition of statistical variables, and for the addition and renaming of fields; it can then be used in Philcarto and in the processing chain to instantiate atlases in uMap. Compared with (Table 1), the output file includes modified, renamed, created, and reordered variables. Suffixes added to variable names enable them to be typed directly by Philcarto [72, 86]. In our database (Table 2), the suffix _R_ indicates a report variable and _N_ indicates a nominal variable. After processing in R, the output file in \u003cem\u003e.xlsx\u0026nbsp;\u003c/em\u003eformat now contains 24 variables.\u003c/p\u003e\n\u003cp\u003eIn the statistical calculation stage\u003cstrong\u003e\u0026nbsp;(\u003c/strong\u003esee in 3.2.3., Table 2, line sections a [25] and a [29]), Excel spreadsheets are supplemented by the additional statistical variables Chi-square and Chi-square with logarithmic transformation \u003cem\u003e\u0026quot;Khi2LnFr\u0026quot;,\u003c/em\u003e the methods of which are described in (section. 3.3.2: Calculation of the four atlas variables). Variables are sometimes renamed (see in 3.2.3., Table 2, line sections b [25] and b [29]) to bring them into line with the Philcarto file header standard: for example, the variable \u003cem\u003e\u0026quot; (Obs - Theo) ^2/Theo_Fr\u0026quot;\u0026nbsp;\u003c/em\u003eor Khi-deux brut is renamed \u0026ldquo;\u003cem\u003eDENSI\u0026rdquo;.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAt the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot;(TrafficLight) classification stage (see in 3.2.3., Table 2, lines [31] and [36]), a new series of Excel processes is used to classify the values of the \u0026quot;\u003cem\u003eKhi2LnFr\u003c/em\u003e\u0026quot; variable to give the fourth \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) atlas variable associated with three colors, and to create the sub-national maps in Philcarto using the \u0026quot;\u003cem\u003eSurmortalit\u0026eacute;_feu_tricolore_umap\u003c/em\u003e\u0026quot;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e(Excess Mortality_TrafficLight_umap)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003evariable. This is also the stage at which the input file \u003cem\u003einsee.xlsx\u003c/em\u003e is\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eenriched with information variables such as\u0026nbsp;\u0026quot;\u003cem\u003eSurmortalite_feu_tricolore_umap\u003c/em\u003e\u0026quot; (Excess Mortality_TrafficLight_umap), recognized directly by the suffix \u0026quot;umap\u0026quot;, which is used to interactively display the communal tooltips\u0026nbsp;(on hover) of uMap atlas instantiations.\u003c/p\u003e\n\u003cp\u003eAt the Philcarto processing stage (see in 3.2.3., Table 2, line sections [31] and [36]), we present the following files as input to the software: \u003cem\u003espreadsheet-insee.xlsx\u0026nbsp;\u003c/em\u003e(France as a whole, ~36,000 communes online) and BaseMap\u003cem\u003e.shp\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003e(georeferenced shapefile for the region or France as a whole) [72, 86]. As an output, we obtain an initial cartography associated with our four atlas variables designated \u003cem\u003e\u0026quot;SMM_10000\u0026quot;, \u0026quot;DENSI\u0026quot;, \u0026quot;Khi2LnFr\u0026quot;\u003c/em\u003e and \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight). The \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) atlas variable is named \u0026quot;\u003cem\u003eSurmortalit\u0026eacute;_feu_tricolore\u003c/em\u003e\u0026quot; (Excess Mortality_TrafficLight). The output formats used are:\u003cem\u003e\u0026nbsp;.ai; .pmc; .emf;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003eand\u003cem\u003e\u0026nbsp;.kml.\u003c/em\u003e The\u003cem\u003e\u0026nbsp;.kml\u003c/em\u003e format is georeferenced with \u003cem\u003ethe Syst\u0026egrave;me G\u0026eacute;od\u0026eacute;sique Mondial (WGS 84, World\u0026nbsp;\u003c/em\u003eGeodetic \u003cem\u003eSystem 1984\u003c/em\u003e) \u0026ndash; \u003cem\u003eEPSG\u003c/em\u003e_4326\u0026nbsp;[29, 31, 105, 106, 107, 108, 109, 110]\u003cem\u003e.\u0026nbsp;\u003c/em\u003eThe Keyhole Markup Language \u003cem\u003eKML\u0026nbsp;\u003c/em\u003eor\u003cem\u003e\u0026nbsp;kml\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eformat\u0026nbsp;\u003c/em\u003eenables interoperability and the start-up of the processing chain in other software environments, with the aim of obtaining a ready-to-use map in .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003eformat for the \u003cem\u003euMapOSM.fr\u0026nbsp;\u003c/em\u003eremote server (see Figure 1 and Figure 3\u003cem\u003e)\u003c/em\u003e. The maps are called choropleth or color range maps and are\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003esystematically accompanied by a legend discretized using Jenks\u0026apos; method (see section 3.3.3) [122, 123].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e2\u003c/strong\u003e \u003cem\u003eINSEE\u003c/em\u003e variables transformed after processing in R and Excel\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"668\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"7.934131736526946%\"\u003e\n \u003cp\u003e\u003cstrong\u003eGroup\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.880239520958083%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLine numbers\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"82.18562874251496%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable names per line\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"7.934131736526946%\" rowspan=\"5\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.880239520958083%\" valign=\"bottom\"\u003e\n \u003cp\u003e[1]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"82.18562874251496%\"\u003e\n \u003cp\u003e\u0026quot;Iden\u0026quot; \u0026quot;Iden_1\u0026quot; \u0026quot;Id_dep\u0026quot; \u0026quot;Id_commune\u0026quot; \u0026quot;Label_commune\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003e[6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Population\u0026quot; \u0026quot;Muni_population\u0026quot; \u0026quot;Dc_2018\u0026quot; \u0026quot;Dc_2019\u0026quot; \u0026quot;Dc_2018_2019\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003e[11]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Dc_2020\u0026quot; \u0026quot;Excess mortality\u0026quot; \u0026quot;Average_age\u0026quot; \u0026quot;SM_10000\u0026quot; \u0026quot;SMM_10000_R_\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003e[16]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;status\u0026quot; \u0026quot;Period\u0026quot; \u0026quot;Id_2020_reg\u0026quot; \u0026quot;Id_2020_dep\u0026quot; \u0026quot;Label_2020_dep\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003e[21]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Label_2020_reg\u0026quot; \u0026quot;Period_N_\u0026quot; \u0026quot;statut_N_\u0026quot;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u0026quot;Excess_mortality_R_\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"7.934131736526946%\" rowspan=\"2\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.880239520958083%\" valign=\"bottom\"\u003e\n \u003cp\u003ea [25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"82.18562874251496%\"\u003e\n \u003cp\u003e\u0026quot;Dc_2020_Theo_Fr\u0026quot; \u0026quot;(Obs - Theo) ^2/Theo_Fr\u0026quot; \u0026quot;Dc_2020_Theo_reg\u0026quot; \u0026quot;(Obs-Theo)^2/Theo_reg\u0026quot;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003ea [29]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Khi2LnFr\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"7.934131736526946%\" rowspan=\"2\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.880239520958083%\" valign=\"bottom\"\u003e\n \u003cp\u003eb [25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"82.18562874251496%\"\u003e\n \u003cp\u003e\u0026quot;Dc_2020_Theo_Fr\u0026quot; \u0026quot;(Obs-Theo) ^2/Theo_Fr\u0026quot; \u0026quot;DENSI\u0026quot; \u0026quot;Khi2LnFr\u0026quot;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003eb [29] \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Dc_2020_Theo_reg\u0026quot; \u0026quot;(Obs-Theo) ^2/Theo_reg\u0026quot;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"7.934131736526946%\" rowspan=\"2\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.880239520958083%\" valign=\"bottom\"\u003e\n \u003cp\u003e[31]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"82.18562874251496%\"\u003e\n \u003cp\u003e\u0026quot;Id_region\u0026quot; \u0026quot;Khi2LnFr\u0026quot; \u0026quot;10_classes\u0026quot; \u0026quot;Class\u0026quot; \u0026quot;TrafficLight\u0026quot;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.731707317073171%\" valign=\"bottom\"\u003e\n \u003cp\u003e[36]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"89.26829268292683%\"\u003e\n \u003cp\u003e\u0026quot;Class_1\u0026quot; \u0026quot;Excess_mortality_TrafficLight\u0026quot; \u0026quot;Excess_mortality_TrafficLight_umap\u0026quot;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eLegend for Table 2:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFour groups (g) and line sections (l):\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eG1, lines 1 to 21: 24 variables obtained from R.\u003c/p\u003e\n\u003cp\u003eG2, lines a25 to a29: variables N\u0026deg; 25 to 29 obtained in Excel.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eG3, lines b25 to b29: variables N\u0026deg; 25 to 30 prepared for Philcarto.\u003c/p\u003e\n\u003cp\u003eG4, lines 31 to 36: variables N\u0026deg;31 to 38 prepared for the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) map, workflows, and tooltips.\u003c/p\u003e\n\u003cp\u003eSuffixes for G1 variables are shown in bold.\u003c/p\u003e\n\u003cp\u003eNote that we have written the Chi2 variable, \u0026quot;Khi2 and Khi2LnFr\u0026quot; in our databases.\u003c/p\u003e\n\u003ch3\u003e3.2.4. IGN base map data engineering\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eUsing files from the \u003cem\u003eInstitut National de l\u0026rsquo;Information G\u0026eacute;ographique et Foresti\u0026egrave;re (IGN,\u0026nbsp;\u003c/em\u003eNational Institute of Geographic and Forest Information) (see Material section 3.1.), we created base maps at the scales of the \u003cem\u003eDROMs/OSDRs\u003c/em\u003e\u003cem\u003e,\u003c/em\u003e large regions (grouped regions) and the arrondissements of Lyon, Paris and Marseille, and georeferenced them \u003cem\u003ethe Syst\u0026egrave;me G\u0026eacute;od\u0026eacute;sique Mondial (WGS 84,\u0026nbsp;\u003c/em\u003eWorld Geodetic System 1984) for \u003cem\u003eGPS\u003c/em\u003e [29, 31]), resulting in a \u003cem\u003e.kml\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003emap output \u003cem\u003efor\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eCAO-DAO/CAMD\u0026nbsp;\u003c/em\u003esoftware (see material section, Figure 1: Synthetic (a) and detailed (b) architectures of uMap atlases for France and Figure 3: Data pipeline, geomatic processing software system - step 1 - without webmapping).\u003c/p\u003e\n\u003cp\u003eFor reasons of vector file size, we have broken down the \u003cem\u003e.shp\u003c/em\u003e shapefile of France\u0026apos;s communes into 4 major sub-national commune regions using the \u0026quot;\u0026Eacute;clats\u0026quot; tool [71, 72]. One or more thicker administrative boundary layers are added to the base map. We produce as many .\u003cem\u003egeojson\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003elayers for uMap as there are choropleth maps produced for an atlas. That is 12 geojson layers for the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) atlas (4 large regions, 3 arrondissements and 5 OverSeas Departments and Regions \u003cem\u003eDROMs/OSDRs\u003c/em\u003e. For further information on map background engineering, please refer to the author\u0026apos;s works\u0026nbsp;[71, 72, 124-126].\u003c/p\u003e\n\u003ch3\u003e3.2.5.\u0026nbsp;Geomatic data engineering\u003c/h3\u003e\n\u003cp\u003eFor the sake of clarity, we present our data pipelines and workflows mainly in schematic form.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGeojson creation step for uMap (step 1 without webmapping)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003e.kml\u0026nbsp;\u003c/em\u003e(direct) file produced by Philcarto is our first file to undergo geomatic processing flows (workflow) via a data pipeline (see uMap environment sections 3.4.1: Step 1- geomatics without webmapping and 3.4.2: Step 2- geomatics with webmapping of uMap environments). In several software packages, transformations made to the \u0026ldquo;direct \u003cem\u003e.kml\u003c/em\u003e\u0026rdquo; result in the final output file in \u003cem\u003e.geojson\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003eformat \u003cstrong\u003e(\u003c/strong\u003esee 3.4.1.1., Figure 3: Data pipeline, geomatic processing software system \u0026ndash; step 1 without webmapping).\u003c/p\u003e\n\u003cp\u003eThe workflow in QGis is based on two input files \u0026ldquo;GE_vscode\u003cem\u003e.geojson\u0026rdquo;\u0026nbsp;\u003c/em\u003eand the \u0026ldquo;spreadsheet \u003cem\u003e.csv\u0026rdquo; (utf8)\u003c/em\u003e, and on the workflow model generated using the \u0026quot;modeler\u0026quot; tool, which generates files in .\u003cem\u003emodel3\u0026nbsp;\u003c/em\u003eformat\u003cem\u003e.\u0026nbsp;\u003c/em\u003eThese models\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eenable batch processing. From QGis, we obtain the output files, in the form of maps and enriched data in .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003eformat, ready for uMap.\u0026nbsp;See, both figures: (Figure\u0026nbsp;4:\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eWorkflow QGis \u003cem\u003e(.model3\u003c/em\u003e) part 1-obtaining the TrafficLight.\u003cem\u003egeoson\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003emap for uMap \u0026ndash; step 1 - without webmapping; Figure 5: Workflow QGis (\u003cem\u003e.model3\u003c/em\u003e) part 2 - batch processing \u0026ndash; TrafficLight.\u003cem\u003egeoson\u003c/em\u003e for uMap- step 1- \u0026nbsp;without webmapping)\u003c/p\u003e\n\u003cp\u003eFor example, the \u0026quot;\u003cem\u003eSurmortalite_feu_tricolore_umap\u003c/em\u003e\u0026quot; (Excess Mortality_TrafficLight_umap) variable is automatically transformed to display structured tooltips when the mouse is moved over the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLigtht) atlas.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLegends creation stage (step 2 with webmapping)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003emap obtained from processing the .kml format (in step 1 without webmapping) is uploaded as an input file \u003cem\u003eto the uMapOSMfr remote server\u003c/em\u003e. Once in place, the \u003cem\u003e.geojson\u003c/em\u003e map is managed using the uMap \u0026quot;layer window\u0026quot; manager and \u0026quot;dashboard\u0026quot; (see in section 4. Results, Figure 11: uMap atlas by TrafficLight, France, \u003ca href=\"http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.572/4.944\"\u003e\u003cem\u003eaddress URL the atlas.\u003c/em\u003e\u003c/a\u003e)\u003c/p\u003e\n\u003cp\u003eThe legend associated with each \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003emap is displayed using the legend creation and formatting data pipeline shown in Figure 7.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe workflow for creating caption thumbnails is shown in Figure 8, while the workflow for formatting captions (automated in Photoshop using \u0026ldquo;script.\u003cem\u003eatn\u003c/em\u003e\u0026rdquo;) is shown in Figure 9.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e.png\u0026nbsp;\u003c/em\u003eor\u003cem\u003e\u0026nbsp;.jpg legend\u0026nbsp;\u003c/em\u003efiles are uploaded to a remote\u003cem\u003e\u0026nbsp;\u003c/em\u003eserver at the University of Lille, as these formats cannot be hosted on the uMap\u003cem\u003eOSM\u003c/em\u003e server, which is reserved exclusively for .geojson files [22 , 23]. (Figure 12: uMap atlas by TrafficLight- zoom on the south-west regions - Bayonne town tooltip. \u003ca href=\"http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.572/4.944\"\u003e\u003cem\u003eURL address of the atlas.\u003c/em\u003e\u003c/a\u003e)\u003c/p\u003e\n\u003ch2\u003e3.3. Step 1 method - geomatics without webmapping\u0026nbsp;\u003c/h2\u003e\n\u003ch3\u003e3.3.1. Choice of scales and meshes\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eFor our mortality atlases, we have chosen to present a France-wide analysis included the OverSeas Departments and Regions \u003cem\u003eDROMs/OSDRs\u003c/em\u003e at a fine territorial grid (commune or arrondissement of the cities of Lyon, Marseille and Paris) and also to complement existing studies that mainly process information at coarser grids [1].\u003c/p\u003e\n\u003ch3\u003e3.3.2.\u0026nbsp;Calculation of the four atlas variables\u003c/h3\u003e\n\u003cp\u003eStatistical calculations are made using death data from the French \u003cem\u003eNational Institute of\u0026nbsp;\u003c/em\u003eStatistics\u003cem\u003e\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;\u003c/em\u003eEconomic Studies\u0026nbsp;(\u003cem\u003eINSEE\u003c/em\u003e)\u0026nbsp;to create atlases (see section 3.2. Methods) based on the following four variables:\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe \u0026quot;\u003cem\u003eSMM_10000\u003c/em\u003e\u0026quot; excess- and under mortality variable\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFormula (a)\u003cstrong\u003e:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/strong\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cdiv id=\"ftn1\"\u003e\n \u003cp\u003eThis variable is calculated from data on communal deaths from all causes, with three reference years: 2018, 2019 and 2020. The numerator (the difference between deaths in 2020 and the average number of deaths in 2018 and 2019) is divided by the municipal population (2017 and per commune); this ratio is then multiplied by 10,000. This variable highlights over- or under-mortalities per 10,000 inhabitants.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eThe \u0026quot;\u003cem\u003eDENSI\u003c/em\u003e\u0026quot; variable for a Raw Chi square (Raw Chi2\u003c/strong\u003e)\u003c/p\u003e\n \u003cp\u003eFormula (b):\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003cbr\u003e\u003cbr\u003e\u003cbr\u003e\u003cbr\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eWe calculate a density of deaths, which is the ratio of the number of national deaths to the total French municipal population (metropolitan or excluding \u003cem\u003eD\u0026eacute;partements et R\u0026eacute;gions d\u0026apos;Outre-Mer\u0026nbsp;\u003c/em\u003e(\u003cem\u003eDROMs/OSDRs,\u003c/em\u003e the OverSeas Departments and Regions)\u003cstrong\u003e)\u0026nbsp;\u003c/strong\u003e(~64M); this ratio is then brought down to the municipal population, enabling a comparison to be made between actual deaths and theoretical deaths. In this way, we can highlight over- or under-mortalities, which are more relevant and significant in terms of probability compared with the national average.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eThe \u0026quot;Chi2LnFr\u0026quot; variable for a Chi square Ln Fr. \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eFormula (d)\u003cstrong\u003e: \u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eUsing a logarithmic scale, the \u0026ldquo;\u003cem\u003eDENSI\u003c/em\u003e\u0026rdquo; variable is transformed with new values that informs directly about levels of excess- or under mortality compared with the national average.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eThe \u0026quot;TrafficLight\u003c/strong\u003e\u0026quot; \u003cstrong\u003evariable\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eThe values of the Chi2LnFr variable or even noted (Khi2LnFr)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eare grouped into three color classes (green, orange, red) using Jenks\u0026apos; discretization method in Philcarto. The semantic interest of the mapped \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) variable provides humans with statistical information that is simplified (three classes) and directly understandable (given the strong symbolism of a traffic light).\u003c/p\u003e\n \u003ch3\u003e3.3.3.\u0026nbsp;Jenks map legend classes\u003c/h3\u003e\n \u003cp\u003eEvery map legend is based on a statistical method for discretizing the data. We have chosen Jenks\u0026apos; method, also known as the \u0026quot;natural threshold method\u0026quot;, which takes as input a defined number of classes. This method maximizes the variance between classes and minimizes the variance within classes, thus making each group of geographical units (communes or arrondissements) more homogeneous (compared to the other methods proposed by Philcarto).\u0026nbsp;Computer calculations can be time-consuming, but the result is optimal. The produced maps are fairly objective, if the number of classes is greater than 5\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eThis method is better suited to multimodal variables possessing a large enough number of individuals to be able to bring out thresholds. It waś developed by G. F. Jenks, an American geographer and cartographer, in the 1960s [127, 128].\u003c/p\u003e\n \u003ch3\u003e3.3.4.\u0026nbsp;Terminal repository - ten national classes, sub-national class bases\u003c/h3\u003e\n \u003cp\u003eA map of the whole of France is created for each variable (three maps), using Jenks\u0026apos; method and a legend of 10 classes. The boundary values of the 10 classes are stored in a repository and used to create thematic maps (of the same variable) on a sub-national scale (region, large region, district \u003cem\u003e(\u0026ldquo;arrondissement\u0026rdquo;)\u003c/em\u003e and the OverSeas Departments and Regions\u003cem\u003e\u0026nbsp;(DROMs/OSDRs\u003c/em\u003e\u003cstrong\u003e)\u003c/strong\u003e.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eE\u003c/strong\u003e\u003cstrong\u003example of ten-class frame of reference\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eSee Table 3 below, showing the boundaries of the 10 classes for the variables \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight), and \u0026quot;Khi2LnFr\u0026quot;. For the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) variable, we essentially target excess mortality compared with the national average. Thus, 3 color classes green, orange, and red represent increasing excess mortality:\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(Green =\u0026gt; 0 - 1.83, means close to the average to 2 times above the average; Orange =\u0026gt; 1.83 - 3.74, means 2 to 4 times higher than the average and Red =\u0026gt; 3.74 - 11.07, means 4 to 11,07 times higher than the average).\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTable 3\u003c/strong\u003e: Ten-class frame of reference based on Jenks discretization (variable: Khi2LnFr)\u003c/p\u003e\n \u003cdiv align=\"center\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"549\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e\u003cstrong\u003eClass\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e\u003cstrong\u003eKhi2LnFr interval\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo. of communes \u003csup\u003e4\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrafficLight\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[+3.736, +11.075]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e581\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003eRed\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[+1.826, +3.736[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e1724\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003eOrange\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[+0.747, +1.826[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e4629\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003eGreen\u003csup\u003e3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-0.074, +0.747[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e7569\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003eGreen\u003csup\u003e3\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-0.844, -0.074[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e8409\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-1.712, -0.844[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e7070\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-3.056, -1.712[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e3466\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-5.215, -3.056[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e976\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-8.78, -5.215[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e340\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"15.300546448087431%\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"32.42258652094718%\"\u003e\n \u003cp\u003e[-21.245, -8.78[\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.51183970856102%\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.76502732240437%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eLegend for table 3\u003c/p\u003e\n \u003cp\u003eFour out of ten classes belong to the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight):\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u003csup\u003e1\u003c/sup\u003eRed [3.74, 11.07], 4 to 11 times higher than average.\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u003csup\u003e2\u003c/sup\u003eOrange [1.83 - 3.74[, 2 to 4 times higher than average.\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u003csup\u003e3\u003c/sup\u003eGreen [0, 1.83[, close to average to 2 times above average.\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u003csup\u003e4\u003c/sup\u003e\u0026quot;communes\u0026quot; or number of towns identified (France as a whole).\u003c/em\u003e\u003c/p\u003e\n \u003ch2\u003e3.4. uMap environment and atlases by variable - two-step geomatics action plan\u003c/h2\u003e\n \u003cp\u003eIn order to obtain an atlas environment (for each variable) instantiated in uMap, we followed a two-step geomatic action plan: \u0026quot;step 1, geomatic processing \u003cem\u003ewithout webmapping\u0026quot;\u0026nbsp;\u003c/em\u003ebased on the construction of maps in .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003eformat \u003cem\u003eoutside uMap\u003c/em\u003e, then \u0026quot;step 2, geomatic processing with webmapping\u0026quot; based on the creation of legends with the help of uMap.\u003c/p\u003e\n \u003ch3\u003e3.4.1.\u0026nbsp;\u0026quot;Step 1- geomatics without webmapping\u003c/h3\u003e\n \u003cp\u003eSee in 3.2.1., Figure 1: Synthetic (a) and detailed (b) webmapping architectures of uMap atlases for France\u003c/p\u003e\n \u003ch4\u003e3.4.1.1. Step 1 data pipeline description\u0026nbsp;\u003c/h4\u003e\n \u003cp\u003eVarious software and processing methods are used to obtain instances of the \u0026ldquo;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026rdquo; (TrafficLight) atlas in uMap.\u003c/p\u003e\n \u003cp\u003eIn order to obtain files in \u003cem\u003e.kml format and then in .geojson format\u0026nbsp;\u003c/em\u003eready for uMap, we processed the information in a software suite producing a data pipeline [22]. To illustrate the complexity of the data pipeline, we present its schematic in Figure 3 below. It shows a succession of tasks ordered in various software packages which are mainly of the following types: \u003cem\u003eCartographie et Dessin Assist\u0026eacute;s par Ordinateur\u003c/em\u003e \u003cem\u003e(CAO-DAO/CAMD\u003c/em\u003e Computer Aided Mapping and Drawing), \u003cem\u003eSIG/GIS\u003c/em\u003e, Globe, source code editors, spreadsheet, etc. (see Material section 3.1.).\u0026nbsp;\u003c/p\u003e\n \u003ch5\u003e\u003cstrong\u003eData pipeline diagram step 1 - from .kml to .geojson for uMap\u003c/strong\u003e\u003c/h5\u003e\n \u003cp\u003eFollowing the progression in Figure 3, we can see that maps in \u003cem\u003e.kml\u0026nbsp;\u003c/em\u003efile format are transformed to obtain the final \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003eformat\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eThe \u003cstrong\u003e.\u003cem\u003ekml\u0026nbsp;\u003c/em\u003e\u003c/strong\u003emap thus undergoes a series of processes in several software environments (Google Earth, VScode and QGis, the \u003cem\u003e.geojson\u003c/em\u003e file editor) until the final \u003cem\u003emap.geojson\u0026nbsp;\u003c/em\u003efile is obtained at the end of \u0026quot;step 1: geomatics without webmapping\u0026quot;. For further information, please refer to Figure 1.\u003c/p\u003e\n \u003cp\u003eSometimes the pipeline is partially automated, otherwise it is fully automated, as in the two QGis workflows Figure 4 and Figure 5 below [61]. The task processing chain is carried out within the \u0026quot;same software environment\u0026quot;. For this example, the \u0026quot;graphical modeler\u0026quot; is used [114] to model the automation of an entire geomatic processing chain, starting with an \u0026quot;input\u0026quot;, before obtaining the output \u003cem\u003e.geojson\u003c/em\u003e file (for uMap). For the creation of a map of France, two files were created from the graphic modeler (\u003cem\u003e.model3\u003c/em\u003e): \u0026quot;file: processing chain\u0026quot; and \u0026quot;file: batch processing\u0026quot; (see in section 3.2.1., Figure 1).\u003c/p\u003e\n \u003ch5\u003e\u003cstrong\u003eQGIS workflow diagrams - processing direct .kml to obtain geojson file for uMap - step 1 without webmapping\u003c/strong\u003e\u003c/h5\u003e\n \u003cp\u003eIn QGis, the workflow is broken down into two parts in Figure 4 and Figure 5 below.\u003c/p\u003e\n \u003ch3\u003e3.4.2.\u0026nbsp;\u0026quot;Step 2- geomatics with webmapping of uMap environments\u003c/h3\u003e\n \u003cp\u003eSuccessively following step 1, the final map\u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003efile obtained is ready for use and becomes the \u0026quot;input file\u0026quot; in step 2 geomatics with uMap webmapping [99].\u0026nbsp;\u003c/p\u003e\n \u003cp\u003ePlease note: The \u003cem\u003e.geojson\u003c/em\u003e file obtained in step 1 may not be imported as is, due to size constraints imposed by the \u003cem\u003eOSM\u003c/em\u003e server administrator. In this case, for certain files, we proceed to lighten them by using the MapShaper tool, enabling us to reduce the size of the geojson file [87, 129] by reducing the definition of vector shapes.\u003c/p\u003e\n \u003ch5\u003e\u003cstrong\u003eStep 2 data pipeline diagram with webmapping - .geojson lite for import into uMap\u003c/strong\u003e\u003c/h5\u003e\n \u003cp\u003eFigure 6 legend: A distinct shape is associated with each of the two types of information: diamond: processing; rectangle:\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003einput-output-result.\u003c/p\u003e\n \u003ch4\u003e3.4.2.2. uMap platform features\u0026nbsp;\u003c/h4\u003e\n \u003cp\u003eOnce the \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003efile has been imported\u0026nbsp;onto uMap\u0026apos;s remote OpenStreetMap (\u003cem\u003eOSM\u003c/em\u003e) server, to address the complexity of processing in step 2, we have again diagrammed the data pipeline shown in section 3.4.2.3., Figure 7: Double caption data pipeline\u0026hellip;)\u0026rdquo; below; see also in section 3.2.1., Figure 1: Synthetic (a) and detailed (b) architectures of uMap atlases for France\u0026rdquo;.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIn the uMap environment\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eThere are few or no lines of code to control the uMap dashboard environment. The map display is set in \u003cem\u003ethe Interface Utilisateur\u003c/em\u003e \u003cem\u003e(IU/UI,\u0026nbsp;\u003c/em\u003ethe User Interface) or dashboard. Once the \u003cem\u003eLocalisateur de Ressources Uniformes (URL\u003c/em\u003e, Uniform Resource Locator) address has been launched in the browser, uMap loads each file of the major regions, arrondissements and \u003cem\u003eDROMs/OSDRs\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ein turn. The final display of the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) map can be viewed at the scale of the whole of France.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCreation of a map settings template: based\u0026nbsp;\u003c/strong\u003eon the first settings made in the dashboard for a map in the \u0026ldquo;FeuTricolore\u0026rdquo; (TrafficLight) atlas \u003cem\u003e(i.e., the parameters for displaying the map, its titles, font and legend),\u0026nbsp;\u003c/em\u003ewe generate a settings template. This template is used to automate and save time when integrating and displaying other variable atlases and their maps in \u003cem\u003e.geojson\u003c/em\u003e format in uMap.\u003c/p\u003e\n \u003cp\u003eOpen data contour and health information layers: provide administrative boundaries and health information. They are uploaded in \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003eformat to the uMap\u003cem\u003eOSM\u003c/em\u003e.fr server. They are managed using the uMap layer manager (see Material section 3.1.).\u003c/p\u003e\n \u003cp\u003eTo host the legends created and associated with each \u003cem\u003e.geojson\u003c/em\u003e map of the major regions, arrondissements and \u003cem\u003eDROMs/OSDRs\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eof the France atlas, the creation of ownership of a hosting server is required. Legends in \u003cem\u003e.jpeg\u0026nbsp;\u003c/em\u003eor \u003cem\u003e.png\u003c/em\u003e format cannot be uploaded to the same remote server as the atlas \u003cem\u003e.geojson\u003c/em\u003e files.\u003c/p\u003e\n \u003cp\u003eThe uMap atlas is secure from the moment it is created, simply set to secure uMap - security maintenance of the \u003cem\u003eOSM\u0026nbsp;\u003c/em\u003eremote server is managed by four administrators from the \u003cem\u003eOSM\u0026nbsp;\u003c/em\u003eassociation, with three servers currently hosted by OVHcloud France. Map data in \u003cem\u003e.geojson\u003c/em\u003e files are not encrypted on the server. The servers are non-compliant with the \u003cem\u003eR\u0026egrave;glement G\u0026eacute;n\u0026eacute;ral sur la Protection des Donn\u0026eacute;es (RGPD/GDPR\u003c/em\u003e, general data protection regulation) [130].\u003c/p\u003e\n \u003cp\u003eWe are not hampered by the few weaknesses of the uMap environment described in the literature by [59, 62] notably that the platform is not very collaborative, which has no impact on our work.\u003c/p\u003e\n \u003ch4\u003e3.4.2.3.\u0026nbsp;Step 2 data pipeline description\u003c/h4\u003e\n \u003cp\u003eTo address the complexity of geomatic processing for atlas embellishment in uMap, we have schematized the data pipeline that creates and associates a double legend to each \u003cem\u003e.geojson\u003c/em\u003e map.\u003c/p\u003e\n \u003ch5\u003e\u003cstrong\u003eDouble caption data pipeline diagram - step 2\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ewith webmapping\u003c/strong\u003e\u003c/h5\u003e\n \u003cp\u003eFigure 7 includes three blocks of procedures numbered from 1 to 8. Following the progression, we start with a file from Philcarto, this time in \u003cem\u003e.emf\u0026nbsp;\u003c/em\u003eformat,\u003cem\u003e\u0026nbsp;\u003c/em\u003ewhich at the end of the chain results in a double map legend with QRcode \u003cem\u003ein .png\u0026nbsp;\u003c/em\u003eformat; these legends are all hosted on a remote server on the univ-lille.fr domain and are associated with each \u003cem\u003e.geojson\u003c/em\u003e map on the uMap\u003cem\u003eOSM\u003c/em\u003e server.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eThumbnail workflow diagram - preparing legends for step 2 uMap webmapping\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cstrong\u003e\u003cem\u003eWorkflow diagram in Photoshop - double legend with and without QRcode -step 2 with webmapping\u003c/em\u003e\u003c/strong\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e[8]The plyr library is replaced by \u0026quot;dplyr (for data blocks)\u0026quot; and \u0026quot;purr (for lists)\u0026quot;.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Results","content":"\u003cp\u003eHere, we mainly present the results of the Philcarto cartographic environment (step 1) and the results of step 2 geomatics with webmapping (see section 3.4.2: Step 2- geomatics with webmapping of uMap environments).\u003c/p\u003e \u003cp\u003eIn step 2, each map \u003cem\u003eof the regions making up the France atlas by variable\u003c/em\u003e is imported into uMap's remote \u003cem\u003eOSM\u003c/em\u003e server. The visual enrichment of the atlas, by displaying legends and tooltips, also requires two processing chains: the first internal, using the uMap dashboard, and the second external, using various environments such as Xn View-XnConvert and Photoshop (see \u003cspan refid=\"Sec13\" class=\"InternalRef\"\u003ematerial\u003c/span\u003e section). In this article, we mainly present the results of the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) atlas and provide the three uMap links to access the online atlases.\u003c/p\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Philcarto geomatics environment (excluding webmapping) - .kml maps \u0026ndash; \u0026ldquo;FeuTricolore\u0026rdquo; (TrafficLight)\u003c/h2\u003e \u003cp\u003eAs a first part of the results, we show below, mainly for step 1, the cartographic results of the Philcarto environment, enabling us to obtain the first exported map in .\u003cem\u003ekml\u003c/em\u003e format (see in section 3.2.1., Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis Philcarto environment consists of a two-part window: the first associates the map result and its legend, and the second, the tools and parameters for map creation.\u003c/p\u003e \u003cp\u003eThen, as the visual aspect of the map remains identical, there is no point in showing it again in the intermediate processing environments, until the final .\u003cem\u003egeojson\u003c/em\u003e format for uMap (reduced in file size) is obtained. In step 1, our initial map is processed using: VSCode, Google Earth, QGis, geojson.io and MapSharper, with only the spreadsheet data enriched at the end.\u003c/p\u003e \u003cdiv id=\"Sec33\" class=\"Section3\"\u003e \u003ch2\u003e4.1.1. Interpreting the FeuTricolore (TrafficLight) choropleth map\u003c/h2\u003e \u003cp\u003e(In Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e: Excess mortality map by traffic light (map of atlas) output.\u003cem\u003ekml\u003c/em\u003e format \u0026ndash; realized with Philcarto), we're looking at a choropleth map and its legend for the \"\u003cem\u003eFeuTricolore\u003c/em\u003e\" (TrafficLight) variable, dressed up with a departmental outline and presented on a French scale. The map covers the period of the 1st covid-19 crisis confinement, from March 1 to May 15 inclusive in 2020.\u003c/p\u003e \u003cp\u003eBased on the Khi2LnFr variable, our statistical indicators are transformed into classes in the legend entitled \"\u003cem\u003esurmortalit\u0026eacute;_feu_tricolore\u003c/em\u003e\" (excess mortality_TrafficLight). There are three classic color classes, for the semantic representation of traffic light, and a pale color (class excluding traffic light) grouping together geographical units close to the average mortality (metropolitan France) or expressing undermortality compared with the average.\u003c/p\u003e \u003cp\u003eThe number of geographical units included in the map is 34833 (6 communes are not included because their population is zero). 20330 communes (around 58%) are classified as not having traffic lights. The reference French municipal population is 63,639,133.\u003c/p\u003e \u003cp\u003eOf the 1,4503 communes classified according to traffic lights (around 42%), 1,2198 communes (around 35%) are in the green class and are positioned in the interval \u003cem\u003e\"close to average to 2 times above average\"\u003c/em\u003e; 1,724 towns (around 5%) are classified as orange, in the range \u003cem\u003e\"2 to 4 times above average\", and\u003c/em\u003e 581 towns (1.66% or around 2%) are classified as red, in the range \u003cem\u003e\"4 to 11 times above average\"\u003c/em\u003e, which represents an extremely significant excess mortality rate.\u003c/p\u003e \u003cp\u003eBelow the classes in the \"\u003cem\u003esurmortalit\u0026eacute;_feu_tricolore\u003c/em\u003e\" \u0026ldquo;excess mortality_TrafficLight\u0026rdquo; legend is a bar chart, with the heights of the rectangles proportional to the number of spatial units (34833) included in each modality of the distributed variable, reading from left to right: 581 (2%); 1724 (5%); 12198 (35%) and 20330 (58%).\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNB\u003c/strong\u003e \u003cp\u003eThe aim of this map is to alert the observer at a glance (visually), and to simplify interpretation for the general public, using semantics known to all.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec34\" class=\"Section2\"\u003e \u003ch2\u003e4.2. uMap geomatics environment (with webmapping) - instantiation of atlas-\u0026ldquo;FeuTricolore\u0026rdquo; (TrafficLight) - OSM remote server\u003c/h2\u003e \u003cp\u003eThe geomatic results obtained in \"step 1\" are used and integrated into the processing chain processes of step 2. Thus, in the second part of the results, we present below, mainly for step 2 uMap, the showcase of uMap functionalities, the results of the display of the \"\u003cem\u003eFeuTricolore\u003c/em\u003e\" (TrafficLight) atlas for the whole of France and a zoom on the south-west region with a tooltip (parameterized in the uMap dashboard) pointing to the commune of Bayonne.\u003c/p\u003e \u003cdiv id=\"Sec35\" class=\"Section3\"\u003e \u003ch2\u003e4.2.1. Feature showcase: ergonomics and usability in uMap\u003c/h2\u003e \u003cp\u003eOur uMap showcase includes our four atlas instances as main outputs. Three \u003cem\u003eURL\u003c/em\u003es provide access to our atlases (one \u003cem\u003eURL\u003c/em\u003e per variable) for the whole of France, including \u003cem\u003eD\u0026eacute;partements et R\u0026eacute;gions d'Outre-Mer (DROMs/OSDRs\u003c/em\u003e, the OverSeas Departments and Regions). There are as many double legends as there are .\u003cem\u003egeojson\u003c/em\u003e layers imported per atlas. Atlas access and display times are relatively short. However, loading into uMap is somewhat slower, due to the display of the complete atlas, which loads .\u003cem\u003egeojson\u003c/em\u003e layers at commune level for the major regions, arrondissements and \u003cem\u003eDROMs/OSDRs\u003c/em\u003e representing around 36,000 communes. Once the atlas is loaded, navigation in the browser is smooth and without slowdown. The dashboard allows choice and configuration of the desired display functionalities and make them available to users from the atlas' uMap interface. If desired, users can contact the atlas designer to obtain the complete data in .\u003cem\u003egeojson\u003c/em\u003e format or a copy of the atlas.\u003c/p\u003e \u003cp\u003eFrom the user interface, several functions are accessible, accompanied by icons and/or menu names: zoom\u0026thinsp;+\u0026thinsp;and zoom -, search for a place name, full-screen view, center the map on location, measure distances, change the background map, map editing link to OpenStreetMap etc., see guide to uMap basic functions [\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e97\u003c/span\u003e, \u003cspan additionalcitationids=\"CR131 CR132\" citationid=\"CR130\" class=\"CitationRef\"\u003e130\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR133\" class=\"CitationRef\"\u003e133\u003c/span\u003e]. Printing atlases requires (1)- access to the data sharing functionality and (2)- retrieval of exported map data in .\u003cem\u003egpx\u003c/em\u003e, and .\u003cem\u003ekml\u003c/em\u003e formats (for other \u003cem\u003eSIG/GIS\u003c/em\u003e or web services such as My\u003cem\u003eOSM\u003c/em\u003eatic [\u003cspan citationid=\"CR134\" class=\"CitationRef\"\u003e134\u003c/span\u003e]) for printing purposes).\u003c/p\u003e \u003cp\u003eFinally, uMap is an interoperable platform that enables the export and printing of maps based on \u003cem\u003eInterface de Programmation d\u0026rsquo;Application\u003c/em\u003e (\u003cem\u003eAPI\u003c/em\u003e, an Application Programming Interface) [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec36\" class=\"Section3\"\u003e \u003ch2\u003e4.2.2. Instantiating the \u0026ldquo;FeuTricolore\u0026rdquo; (TrafficLight) atlas - uMap world map via OSM\u003c/h2\u003e \u003cp\u003e \u003cb\u003eThe uMap environment in\u003c/b\u003e Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e, \u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e \u003cb\u003econsists of a three-part window\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe first part combines the visual aspect of the France atlas with several tool-buttons on the left of the screen, including our main one for managing and displaying layers (or uploaded regional .\u003cem\u003egeojson\u003c/em\u003e maps), as well as searching for data via the \"Browse data\" link.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe second part of the screen, on the right, is the legend tab, where the legends for each layer in the atlas are integrated.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe third part represents the window banners displaying the map title, designer and properties [\u003cspan citationid=\"CR131\" class=\"CitationRef\"\u003e131\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eHere, for the period of the first French confinement, we see the \"\u003cem\u003eFeuTricolore\u003c/em\u003e\" (TrafficLight) atlas, dressed with departmental and regional outlines and legends associated with regional .\u003cem\u003egeojson\u003c/em\u003e layers on a remote server. In this way, the \"\u003cem\u003eFeuTricolore\u003c/em\u003e\" (TrafficLight) atlas not only displays excess mortality information thanks to the colors displayed on the map and map legend, but also shows structured information in tooltips set from the dashboard.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHere are the \u003cem\u003eURL\u003c/em\u003es for viewing the atlases of the \"\u003cem\u003eSMM_10000\u003c/em\u003e\" variable: \u003cem\u003eatlas URL\u003c/em\u003e and the \"\u003cem\u003eDENSI\u003c/em\u003e\" variable: \u003cem\u003eatlas URL.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cp\u003eThe division of tasks between authors gave us the opportunity to invest in various geomatics and IT development environments for the creation and visualization of our atlases.\u003c/p\u003e\n\u003cp\u003eWe will discuss some of the main points of the two-step geomatics action plan without and with webmapping, such as the data engineering carried out in step 1 on the collection, cleaning and ordering of death data, and such as the processing flows and optimization for the creation of legends in step 2.\u003c/p\u003e\n\u003ch2\u003e5.1. Discussion - INSEE death data engineering\u003c/h2\u003e\n\u003cp\u003eThree selections were made from the death data sets of theFrench National Institute of Statistics and Economic Studies \u003cem\u003e(\u003c/em\u003e\u003cem\u003eINSEE\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003e1-Our first reference is the 2017 census population, which aggregates data from 2015 to 2019. Explanation: communal censuses, managed by \u003cem\u003eINSEE\u003c/em\u003e, take place every year on a fifth of communes with fewer than 10,000 inhabitants and every year by a survey on a sample of 8% of the population for communes with more than 10,000 inhabitants [135]. The population for a reference year N aggregates the data for the five years N-2, N-1, N, N+1 and N+2. Finally, \u003cem\u003eINSEE\u003c/em\u003e indicates that the error induced by the survey method is less than 0.02%. The reference population (2017) chosen for the study aggregates data from 2015 to 2019.\u003c/p\u003e\n\u003cp\u003e2- The municipal population was chosen for the study. Explanation: two types of population census are determined: \u003cem\u003ethe municipal population,\u0026nbsp;\u003c/em\u003ewhich essentially corresponds to usual places of residence, and \u003cem\u003ethe population \u0026quot;counted separately\u0026quot;\u003c/em\u003e, which also includes other types of residence (second homes, students, etc.)\u0026nbsp;[136]\u0026nbsp;so to be able to compare population data from one year to the next without double counting: i.e. without counting duplicates in the population.\u003c/p\u003e\n\u003cp\u003e3- We have approximated the average age of death in the face of missing values. Explanation: some days or months of birth are missing (0.7%). However, this average age could be refined by defining a month and a day of birth (6\u003csup\u003eth\u003c/sup\u003e month of the year for the month and 15\u003csup\u003eth\u003c/sup\u003e day of the month for the day). Our approximation is considered acceptable for the present study, given the low percentage of incomplete data.\u003c/p\u003e\n\u003ch2\u003e5.2.\u0026nbsp;Discussion of the choice of BD data management methods in the RStudio environment\u003c/h2\u003e\n\u003cp\u003eIn\u0026nbsp;theRStudio\u003cem\u003eEnvironnement de D\u0026eacute;veloppement Int\u0026eacute;gr\u0026eacute; (IDE\u003c/em\u003e,integrated development environment), Rmarkdown is an integrated module interwoven with source code (Iterate Programming), allowing users to build the logic followed for at least part of the processing chains in step 1. An example of a \u0026ldquo;summary program\u0026rdquo; that processes the input file \u003cem\u003einsee_data_pop.csv\u0026nbsp;\u003c/em\u003e(see in sections 3.2.1. and 3.2.2., Figures 1, 2 and Table 1) and outputs a ready-to-use file in \u003cem\u003e.xlsx\u003c/em\u003e and \u003cem\u003e.txt\u0026nbsp;\u003c/em\u003eextensionsfor mapping in the Philcarto Computer Aided Mapping and Drawing \u003cem\u003eCAO-DAO/CAMD\u003c/em\u003e software, and for obtaining a \u003cem\u003e.kml\u003c/em\u003e file, can be found in the article\u0026apos;s SF1_supplementary_file_1.zip [70]\u003cem\u003e.\u003c/em\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe recent package for R \u0026quot;Cartography\u0026quot; renamed \u0026quot;mapsf\u0026quot; [122, 137] (not used in the present study), makes it possible to attempt in the future reproducibility of the entire step 1, i.e. this time from the \u003cem\u003einsee.fr\u003c/em\u003einput file to the final output file in \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003eformat, which represents a choropleth map of the atlas ready for import into uMap.\u003c/p\u003e\n\u003cp\u003eIf step 1 can be reproduced in the RStudio environment only, to obtain maps in \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003eformat, then the benefits will be manifold, with the help of the\u0026nbsp;RMarkdown\u0026nbsp;report (which enables R programs to be restarted and initial results to be obtained with a single click; or to more quickly/easier correct any statistical errors in the program).\u003c/p\u003e\n\u003ch2\u003e5.3. Discussion of the choice of the uMap webmapping environment\u003c/h2\u003e\n\u003cp\u003eWe chose an open-source software package that is powerful, easy to use and compatible with other tools, thanks to the interoperability of the software\u0026apos;s input and output file formats. For example, uMap supports the download of data in \u003cem\u003e.umap, .geojson\u003c/em\u003e(OpenStreetMap proprietary format) and \u003cem\u003e.\u003cstrong\u003ekml\u0026nbsp;\u003c/strong\u003e\u003c/em\u003e(Google proprietary format standardized by the Consortium G\u0026eacute;ospatial Ouvert (\u003cem\u003eOGC, Open\u0026nbsp;\u003c/em\u003eGeospatialConsortium) in 2008) formats [1, 59, 99, 105, 131]. It is an ethical environment that does not reuse our own \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003edata loaded on uMap if the sharing rights are not active [62]. uMap is based on Django [138, 139].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe uMap application is fairly simple and intuitive to use, with a dashboard for managing and displaying layers, and numerous integrated functions and resources (world map-type maps) available [130, 132, 133, 140, 141]. We have not encountered any bottlenecks in using the platform. Extensive documentation is available online [131, 132, 138]. The software package is maintained by a dynamic community of developers and geomaticians[9] [62, 97, 99, 100, 142, 143].\u003c/p\u003e\n\u003cp\u003euMap is used in many areas of collaborative cartography. The current uMap environment and its remote server managed by OpenStreetMap.fr (\u003cem\u003eOSM\u003c/em\u003e) hosts maps from all over the world. Using keywords such as \u0026quot;health, mortality\u0026quot;, we can find thematic maps if they are shared with the community (sharing enabled). We have not found, if they exist, choropleth maps by filtering by keyword such as \u0026quot;health\u0026quot; and \u0026quot;mortality\u0026quot; [144].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSharing our mortality atlases on the first French containment in 2020, could be attractive both in terms of the theme and the semantic interest of the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) atlas to the instantaneous message of excess mortality. Publishing our atlases[10] in uMap, with authorization to share and export data, will be a good test to evaluate the potential of shared information and feedback from the uMap community and platform.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIt is worth noticing that uMap offers functionalities not yet used for our atlases\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) The \u0026quot;\u003cem\u003eSlideshow\u0026quot;\u0026nbsp;\u003c/em\u003esee example in [133],\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(2) The \u0026quot;\u003cem\u003eIntegrate and share a uMap\u0026quot;\u0026nbsp;\u003c/em\u003eicon is a feature that allows the map to be integrated within a website, and offers eight export options, including the open standard \u003cem\u003eGPS\u003c/em\u003e exchange format (Global Positioning eXchange\u003cem\u003e\u0026nbsp;GPX\u003c/em\u003e) standardized by the \u003cem\u003eOGC\u003c/em\u003e for the exchange of \u003cem\u003eGPS\u003c/em\u003e coordinates [31, 145]. This sharing function provides a short web address for the map\u0026apos;s \u003cem\u003eLocalisateur de Ressources Uniformes (URL\u003c/em\u003e, Uniform Resource Locator) resources.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(3) Finally, the uMap environment can be installed locally, which is ideal for a server dedicated exclusively tohealthcare.Additional IT developments and new functionalities can be added to enrich the local uMap environment. \u0026ldquo;uMap\u0026rdquo; provides an \u003cem\u003eInterface de Programmation d\u0026rsquo;Application\u003c/em\u003e (\u003cem\u003eAPI,\u003c/em\u003e an Application Programming Interface) based on Python and JavaScript languages [138].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(4) Dynamic updating of \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003efiles is possible\u0026nbsp;[146].\u003c/p\u003e\n\u003ch2\u003e5.4. Discussion of two-step geomatic methods for the uMap environment\u003c/h2\u003e\n\u003cp\u003eThe vast majority of maps produced in the humanities and social sciences are the result of sometimes highly complex processes, involving numerous processing software (or systems) and file formats with limited possible automation of tasks. This is the case in our work, where through our task processing stages, we have been confronted with a pipeline of complex, non-interconnected data that weighs down processing and final execution times for obtaining atlases in uMap.\u0026nbsp;However, even if there is no interconnection between systems\u0026nbsp;[147], software programs produce file formats that enable interoperability when moving from one software environment to another.\u003c/p\u003e\n\u003cp\u003eBelow we present a critical analysis of our two-step geomatics action plan with and without webmapping.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e5.4.1.\u0026nbsp;Discussion of two-step geomatics action plan with and without webmapping\u003c/h3\u003e\n\u003cp\u003eIn view of the results obtained, our action plan offers clear advantages, since we use a large proportion of licensed open-source software, free software, licensed open data, as well as environments and functionalities adapted to our needs, that have enabled us to fully automate certain tasks.\u003c/p\u003e\n\u003cp\u003eHowever, despite these advantages, our two-step geomatic action plan is rather cumbersome, as it relies on a substantial number of geomatic processes. We list below at least four of the main drawbacks encountered:\u003c/p\u003e\n\u003cp\u003e(1) lack of access to the latest versions of proprietary software makes task automation sub-optimal. This is the case, for example, with our (old) version of Adobe Photoshop 7.0 [92];\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(2) due to a lack of geomatic functionality, the data pipeline is not automatically interconnected: an example is shown in the following article\u0026nbsp;[147]\u0026nbsp;which looked at a company\u0026apos;s system that relied on a PostGIS database, Geoserver and OpenLayers visualizations on a PHP server;\u003c/p\u003e\n\u003cp\u003e(3) the function for managing and creating double legends, to be associated with \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003elayers (maps without legends), is not currently implemented to choropleth thematic maps in uMap: this function is expected in the future.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(4) a multitude of .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003emaps has to be produced to instantiate the atlases (by variable) in uMap.\u003c/p\u003e\n\u003ch3\u003e5.4.2. Discussion of design in uMap \u0026nbsp;\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eIn terms of design benefits\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) uMap is licensed under \u003cem\u003eLicence Publique Faites en ce que Vous Voulez (LPFVV/ WTFPL,\u0026nbsp;\u003c/em\u003eDo What The Fuck You Want To Public License) [98] which characterizes it as free of redistribution and modification,\u003c/p\u003e\n\u003cp\u003e(2) As it stands, this software package delivers the expected results without having to worry about security maintenance of the remote server hosting the \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003efiles. The latter is managed by OpenStreetMap or OSM.fr and is an important point for the instantiation of our client-side atlases.\u003c/p\u003e\n\u003cp\u003e(3) We have facilitated the semantic interpretation of mortality by creating the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) atlas, which is considered to be the best messenger to alert us to the excess mortality in the communes and the three arrondissements, compared with the other three atlases of the \u003cem\u003e\u0026ldquo;SMM_10000\u0026rdquo;, \u0026ldquo;DENSI\u0026rdquo; and \u0026ldquo;Khi2LnFr\u0026rdquo;\u0026nbsp;\u003c/em\u003evariables.\u003c/p\u003e\n\u003cp\u003e(4) As for the atlas on the \u003cem\u003eSMM_10000\u0026nbsp;\u003c/em\u003evariable, this indicator of excess mortality per 10000 inhabitants helps highlighting differences in small towns, which make up the majority of the French territory.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe design limitations are as follows\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) The geomatic processing flows (DB management, statistics, variable formatting, etc.) required to obtain ready-to-use \u003cem\u003e.geojson\u0026nbsp;\u003c/em\u003elayers in uMap are cumbersome, complex and not currently reproducible in a single environment. Indeed, the main problem is that our architecture resembles a pipeline system of non-interconnected data, requiring a substantial number of geomatic software tools and several workflows for processing tasks (of which only 2 workflows are fully automated), as in the example of the QGis modeler (see sections 3.2.5: Geomatic data engineering, 3.4.1.1.: step 1: data pipeline description).\u003c/p\u003e\n\u003cp\u003e(2) The legend management system for choropleth maps in color ranges in the uMap platform is not adapted and would require specific development in uMap.\u003c/p\u003e\n\u003cp\u003e(3) Finally, hosting double legends in \u003cem\u003e.png\u0026nbsp;\u003c/em\u003eor \u003cem\u003e.jpg format\u0026nbsp;\u003c/em\u003erequires another remote server (private and university). This is a constraint of the uMap dashboard, that doesn\u0026apos;t anymore allow map legends to be imported in the same place as geojson files.\u003c/p\u003e\n\u003ch3\u003e5.4.3.\u0026nbsp;Discussion on the use of the uMap collaborative platform\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eBenefits of using uMap\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) We have diverted uMap\u0026apos;s primary function, which is considered a geomatic environment for collaborative map sharing, to use uMap as a support for our mortality atlases [59]. Indeed, the publication of thematic atlases using choropleth maps is unusual in uMap. There are many public examples of shared maps on various themes (sports routes, itineraries with hotspot symbolism etc.).\u003c/p\u003e\n\u003cp\u003e(2) The France-wide display of the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) atlas in uMap is highly satisfactory, as it allows instant interpretation of excess mortality alerts.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(3) We enhance the appeal of maps with a fairly fine communal grid, so that users can analyze in detail the areas where they live and work (one or more communes).\u003c/p\u003e\n\u003cp\u003e(4) We have not encountered any technical problems displaying instantiated atlases in different web browsers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(5)\u0026nbsp;There is no security maintenance on the uMapOSM.fr server hosting the geojson layers.\u003c/p\u003e\n\u003cp\u003e(6) The possibility of creating a parameter template is offered by map cloning and has been used.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMain limitations in terms of use\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) Atlases cannot be grouped within one single uMap \u003cem\u003eURL\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e(2) An incident with the uMapOSM server at the end of May 2023 caused access errors to all our atlases. This problem was reported on the \u003cem\u003eOpenStreetMap Project\u0026nbsp;\u003c/em\u003eforums and GitHub\u0026nbsp;[99]\u0026nbsp;and resolved fairly quickly.\u003c/p\u003e\n\u003ch2\u003e5.5. Discussion based on the example of workflow optimization for double captions and the limits encountered (step 2)\u003c/h2\u003e\n\u003cp\u003eWe have succeeded in optimizing certain workflows aimed at obtaining .\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003elayers in QGis (step 1) and at creating and managing double legends (step 2). Here we describe the optimization of double legends and the limitations encountered.\u003c/p\u003e\n\u003cp\u003eWe have succeeded in lightening some of the geomatic processing involved in obtaining double legends, thanks to the \u003cem\u003e.atn\u003c/em\u003escripts (proprietary Adobe Photoshop\u0026copy; (PS7) format) that make up our workflow models.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003e.atn\u003c/em\u003escripts are obtained from our old PS7 version by recording macros. It\u0026apos;s a semi-automated procedure that speeds up the processing of captions, while checking that their formatting, such as resized size, is displayed. In consequences, the .\u003cem\u003eatn\u0026nbsp;\u003c/em\u003escript enabled us to save considerable time in processing double captions without and with QRcode. In fact, we avoided manual input errors on multiple, repetitive tasks. Ideally, we would like to work with a more recent version of Photoshop, or move on to other non-proprietary, or free, under certain conditions, drawing software.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConcerning the limitations encountered with Photoshop 07 (PS7)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(1) .\u003cem\u003eatn\u0026nbsp;\u003c/em\u003escripts can be further optimized using batch processing, currently impossible to realize without bug in the PS7.\u003c/p\u003e\n\u003cp\u003e(2) We noted a problem of incomplete macro recording using the mouse or \u003cem\u003eInteraction Homme Machine\u003c/em\u003e (\u003cem\u003eHMI/HMI,\u003c/em\u003e Human Machine Interaction), i.e. the actions recorded in the .atn file. To get around this problem, editing .\u003cem\u003eatn\u0026nbsp;\u003c/em\u003escripts in \u003cem\u003e.xml\u003c/em\u003e formatusing existing JavaScript utilities would be a solution worth testing. This would require working on a virtual machine with a suitable older Windows XP operating system for existing utilities [115].\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e5.6. Discussion on the proliferation of geomatics tools, languages and tips\u003c/h2\u003e\n\u003cp\u003eSoftware evolution is rapid in IT and geomatics: it is advisable to carry out an organic (dynamic) watch at the beginning and throughout the project [116].\u003c/p\u003e\n\u003cp\u003eAll\u0026nbsp;Computer Aided Mapping and Drawing\u003cem\u003e\u0026nbsp;CAO-DAO/CAMD\u003c/em\u003e software used in geomatics, as well as webmapping environments such as collaborative and non-collaborative platforms, require in-depth investment (during the project, with the help of user guides) [124, 148]), a dedicated geomatics forum, and exchanges with IT specialists. These environments generally combine programming languages (Python for scripting in QGis), scripting languages (Scheme) [117] and Visual Basic for Application\u003cem\u003e\u0026nbsp;(VBA)\u003c/em\u003e and Philcarto macro languages.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGenerally speaking, we use the basic functionalities available in geomatics environments as well as utilities made available specifically for an environment such as QGis. For example, for task automation in QGis, one could use the \u0026quot;modeler\u0026quot; tool and its associated \u003cem\u003e.model3\u003c/em\u003efile format for building workflow models.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRegarding these languages (Python, Scheme, JavaScript etc.) and open, free, or proprietary tools for automating tasks or interconnecting systems\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWith the aim of (1) optimizing repetitive tasks as much as possible in the course of our needs and projects, or (2) interconnecting systems in a data pipeline\u0026nbsp;[147]: a good compromise for the geomatician is to invest in these geomatic-computing languages and tools in parallel and on an ongoing basis. An intellectual investment is clearly necessary in order to avoid the waste of time and energy caused by the tedious manual execution of repetitive tasks.\u003c/p\u003e\n\u003ch2\u003e5.7. Discussion of atlas results\u0026nbsp;\u003c/h2\u003e\n\u003ch3\u003e5.7.1. Atlas \u0026quot;Feu Tricolore\u0026quot; (TrafficLight) semantic relevance and scientific openness\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eTraditionally, maps can be difficult to read for people lacking the prerequisites of thematic cartography, or the concepts of graphic semiology. Our atlas has taken this difficulty into account, offering maps aimed at the general public, healthcare decision-makers and political decision-makers who need relevant, more global information that can be communicated to all. The principle of intuitive comprehension is essential if we are to capture and hold users\u0026apos; attention. In particular, among the instantiated atlases, the one presenting the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) variable is clear for all to see, as it provides an effective account of the given situation for each commune in the national territory.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e5.7.2. Atlas Feu Tricolore (TrafficLight)- its warning levels\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eEven though this project is an epidemiological research endeavor, the results of excess mortality in 2020 compared with the two reference years (2018 and 2019) will enable us to examine the causes of excess mortality in certain communes and under mortality in others. In fact, this is one of the interests of our work: to provide this data at a very fine mesh (the commune) and on the scale of the whole of France, including the OverseasDepartmentsandRegions \u003cem\u003eDROMs/OSDRs\u003c/em\u003e, enabling each commune to position itself in relation to its region, department or other administrative entity, or in relation to the national level (as in this article).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e5.7.3.\u0026nbsp;Indirect epidemiological approach: mapped communal grid and INSEE database\u003c/h3\u003e\n\u003cp\u003eThere are two epidemiological aspects of interest that emerge indirectly from our geomatic environment of all-cause mortality atlases for the year 2020.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe first innovative aspect of our study is the availability of statistical information on deaths from all causes at the municipal level, which by definition is less biased than the one obtained with a more aggregated grid (see section. 2.1.2.1: Cartographic information bias\u0026hellip;), and which complements the aggregated cartographic information published on the covid-19 pandemic [1, 37, 149, 150]. The commune is the smallest spatial unit for which \u003cem\u003eINSEE\u003c/em\u003e death data are available, with the exception of large cities (Marseille, Lyon and Paris), for which information is available at the arrondissement level.\u003c/p\u003e\n\u003cp\u003eTwo additional limitations should also be noted, one relating to discretization thresholds and the other to\u0026nbsp;singular cases (small communal population), either a bias in mortality figures\u003cem\u003e.\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe second aspect concerns the interest of the source of the death data itself, and our approach, like that realized of French TV channel, \u003cem\u003eFrance 3 in the Nord - Pas-de-Calais\u003c/em\u003e region, of comparing mortality with previous years and with the communal grid. France 3 stated: \u0026ldquo;Even if the cause of death is not mentioned in the \u003cem\u003eINSEE\u003c/em\u003e data, an excess of mortality, compared with the two previous years, can be an indication of the impact of the coronavirus, whose information h has been aggregated at departmental level. This is how we were able to measure, for example, the impact of the deadly heatwave in France in the summer of 2003 [50]\u0026rdquo;.\u003c/p\u003e\n\u003cdiv id=\"ftn1\"\u003e\n \u003cp\u003e[9] Support on Github, on the Gitter forum which meets once a month (last Thursday of the month) and on the \u003cem\u003eOSM\u003c/em\u003e forum for uMap\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"ftn2\"\u003e\n \u003cp\u003e[10] Publication of the atlas, which is accessible to all, requires the atlas URL to be made available.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"6. Perspectives","content":"\u003cp\u003eWe are to keep in mind that reproducibility of research remains our guideline also for our optimization, implementation and development perspectives for atlases in uMap [121, 151].\u003c/p\u003e\n\u003ch2\u003e6.1. Finalization of atlases for four periods\u0026nbsp;instead of one in a 2\u003csup\u003end\u003c/sup\u003e future version\u003c/h2\u003e\n\u003cp\u003eIn the short term, we intend to supplement our four atlases, including the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot; (TrafficLight) atlas (from the first containment of 2020), with those of a 2\u003csup\u003end\u003c/sup\u003e version for each of the four periods and for the entire year 2020. The periods, selected as a result of our 2020 watch, correspond to the following four political periods: from January 1st to February 28th inclusive (excluding the covid-19 crisis); from March 1st to May 15th inclusive (1st containment covid-19 crisis); from May 16th to October 28th inclusive (excluding the covid-19 crisis containment); and from October 29th to December 31st inclusive (2nd light containment covid-19 crisis). Four periods would give a better analysis of the phenomenon - i.e. a reflection of the policy over the allotted time. It should be remembered that in this article we present only the \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot;\u0026nbsp;(TrafficLight)\u0026nbsp;atlas for the first containment period of 2020.\u003c/p\u003e\n\u003cp\u003eWe have initiated a first statistical exploration on a new extraction of death data from \u003cem\u003eINSEE\u003c/em\u003e on July 23rd, 2021, and have noticed that the methods used to collect deaths differ between our current version (see Material section 3.1.) and a 2nd version planned over four periods. We already know from the statistics for the 2\u003csup\u003end\u003c/sup\u003e version that 13944 communesout of 34,833[11] (i.e. 40% of communes) are concerned by an \u0026ldquo;excess mortality alert\u0026rdquo; using our \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u0026nbsp;\u003c/em\u003e(TrafficLight)\u0026nbsp;variable (2\u003csup\u003end\u003c/sup\u003e version \u0026ndash; 1\u003csup\u003est\u003c/sup\u003e period in the year 2020). Reminder the 1\u003csup\u003est\u003c/sup\u003e version, \u0026ldquo;see section 4.1.1.: Interpreting the TrafficLight choropleth map\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003eSecondly, we also plan to deepen the analysis by coupling the atlas results of the four variables and periods with the health policies jointly implemented on the French national territory. This would enable us to draw up a balance sheet of the efficient policy measures taken (in terms of mortality decline during the year 2020) and those with a more nuanced or questionable impact (in terms of non-significant growth and decline).\u003c/p\u003e\n\u003cp\u003eTo improve our understanding of the communal landscape, following alerts of excess- or under mortality, we could consider cross-referencing the data with the communes\u0026apos; socio-demographic data.\u003c/p\u003e\n\u003cp\u003eIt might also be useful to compare results from France with those from a neighboring country such as Belgium; we are planning to collaborate with a Belgian team, but there are other methodological obstacles to consider. Our different time periods (containment and de-containment) do not strictly correspond; in France and Belgium, for example, the periods (P0, P1, P2 and P3) are partly out of sync [152]. Similarly, it is not certain that in Belgium, the years 2018 and 2019 are reference years, and this should be heeded in the comparative process.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFinally, despite some differences between France and Belgium, the methodology applied in this article is transferable to any geographical area for which mortality data (all causes combined) are available.\u003c/p\u003e\n\u003ch2\u003e6.2. Geomatics: focus on optimization, automation and reproducibility in R\u003c/h2\u003e\n\u003cp\u003eAt the geomatics level, intra- and extra-webmapping processing chains (respectively map production (.\u003cem\u003egeojson\u0026nbsp;\u003c/em\u003elayers\u003cstrong\u003e\u003cem\u003e)\u0026nbsp;\u003c/em\u003e\u003c/strong\u003efor the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) atlas for uMap), and \u0026quot;legends and tooltips\u0026quot; information in uMap need to be optimized. We also intend to optimize processing between various environments to interconnect them. To achieve this, we could call on (1) an \u003cem\u003eExtraire, Transformer et Charger\u003c/em\u003e (\u003cem\u003eETL,\u003c/em\u003e Extract, Transform, Load) pipeline architecture from proprietary software, under various types of license; (2) a programming language such as Python, where libraries exist for most of the set objectives [147, 153 ].\u003c/p\u003e\n\u003cp\u003eOur short- to medium-term goal is to rapidly produce or update atlases instantiated in uMap.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the interests of greater reproducibility, we would also like to focus on a single environment for much of our geomatics processing. To this end, we would like to focus more on the R project environment and its new packages dedicated to cartography[12], i.e. to integrate the entire cartographic processing process rather than working in Excel, Philcarto, Google Earth, VScode and QGis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt is a challenge we believe is achievable: thanks to digital evolution and the interest of developers in geomatics, who are simplifying the work of the humanities and social sciences by simplifying the geomatics processing chain and developing new mapping packages that comply with the rules of the art of graphic semiology.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThese developers invite us to use the same software environment, and to take advantage of the R project to carry out all data processing chains right through to cartographic results. In this respect, these developers\u0026nbsp;[123, 151]\u0026nbsp;invite us to join the concept of reproducible cartography.\u003c/p\u003e\n\u003cp\u003eThus, the \u0026quot;Cartography\u0026quot; package, in its new version renamed \u0026quot;mapsf\u0026quot; more user-friendly, lighter and more robust [123, 137, 154] enables all four methodological steps \u003cem\u003e(data collection; data cleaning and ordering; data analysis and spatial representation\u003c/em\u003e) to be followed and carried out within the same software environment to produce maps, and avoids any dispersal or complication of tasks. In other words, it enables a unified workflow and map reproducibility process.\u003c/p\u003e\n\u003cp\u003eThe packaged R project is a very good choice for \u003cem\u003estep 1 statistics and geomatics without web-mapping\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNB:\u003c/strong\u003e Reproducible cartography raises a stumbling block in the face of sensitive, regulated or even inaccessible health data. Indeed, we could not make health data available for any cartographic reproduction by others, even for a designated community belonging to the healthcare environment. This first barrier is reinforced for a non-healthcare community.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNevertheless, new positive perspectives are emerging thanks to \u003cstrong\u003e\u003cem\u003e\u0026quot;think tanks\u0026quot;\u0026nbsp;\u003c/em\u003e\u003c/strong\u003emade up of healthcare professionals\u0026nbsp;[155]\u0026nbsp;who are working to improve the reusability of healthcare data, particularly hospital data, for public health and epidemiological research.\u003c/p\u003e\n\u003ch2\u003e6.3. Targeted conceptual and geomatic developments uMap\u003c/h2\u003e\n\u003cp\u003eConceptual evolution: to give more meaning to our atlases like \u0026quot;\u003cem\u003eFeuTricolore\u003c/em\u003e\u0026quot;\u0026nbsp;(TrafficLight), for which the uMap platform is originally synonymous with a\u0026nbsp;geomatics environment for collaborative map sharing, we want to share our atlases with the uMap community, with rights open to user interaction. These atlases of all-cause mortality will also be open to personalized comments [59, 133].\u003c/p\u003e\n\u003cp\u003eWe hope to be able to evaluate the interactions and their impact on the attractiveness of the atlases.\u003c/p\u003e\n\u003cp\u003eAll-cause mortality atlases at the local level, fed by an open \u003cem\u003eINSEE\u003c/em\u003e data source, can join this collaborative map-sharing concept without constraint.\u003c/p\u003e\n\u003cp\u003eGeomatics evolution: uMap\u0026apos;s designer has given us the option of installing and developing a customized uMap environment as a basis for implementing various \u0026quot;health atlases\u0026quot;. \u0026ldquo;uMap\u0026rdquo; would then be implemented on a university server where our atlas legends are currently hosted. During the implementation phase, uMap contributors will be able to contribute to the development of a public[13] fork for the creation and management of legends and the display of tooltips [99, 138]. Beforehand, we plan to present the schematics of (Figure 3 and Figure 7) so that contributors can better assess the complexity of the data pipelines of the two geomatic steps and consider an improvement for the creation of legends in uMap.\u003c/p\u003e\n\u003cp\u003eIn addition, this \u0026quot;\u003cem\u003euMap sant\u0026eacute;\u003c/em\u003e\u0026quot; (uMap in health) could be opened up to our partners and/or other scientists. However, it should be reminded that health data is regulated differently from \u003cem\u003eINSEE\u003c/em\u003e data. Within this constrained framework of access and respect for the use of health data, the development of this \u003cem\u003e\u0026quot;uMap sant\u0026eacute;\u003c/em\u003e\u0026quot; (uMap in health)\u0026nbsp;server would be legitimate, but subject to the regulations in force. We could also draw inspiration from the Pixacare medical photo library, with a view to developing and diversifying the use of \u0026quot;\u003cem\u003euMap sant\u0026eacute;\u003c/em\u003e\u0026quot; (uMap in health) or collaborative health map libraries [156].\u003c/p\u003e\n\u003cdiv id=\"ftn1\"\u003e\n \u003cp\u003e[11] Six communes are excluded from the \u0026ldquo;\u003cem\u003eFeuTricolore\u0026rdquo;\u003c/em\u003e (TrafficLight) variable, as their population is zero. They are identified by their local \u003cem\u003eINSEE\u003c/em\u003e codes: 55039-Beaumont-en-Verdunois; 55050-Bezonvaux; 55139-Cumi\u0026egrave;res-le-Mort-Homme; 55189-Fleury-devant-Douaumont; 55239-Haumont-pr\u0026egrave;s-Samogneux and 55307-Louvemont-C\u0026ocirc;te-du-Poivre.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"ftn2\"\u003e\n \u003cp\u003e[12] The oldest spatial packages are represented by: RGDAL: interface between R and GDAL libraries; sp: classes and methods for geolocalized data; rgeos: spatial calculations of the following types: area, buffer zone, intersection, superposition, merge, dissolve. The sf package encompasses the 3 previous packages and provides new, simpler-to-use functionalities, as well as being pipe and operator compatible.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"ftn3\"\u003e\n \u003cp\u003e[13] Software created or modified from existing source code.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"7. Conclusions","content":"\u003cp\u003eWe have demonstrated the value of the uMap \u003cem\u003eGratuit/Libre et Logiciel de Source Ouverte (GLLSO/FLOSS\u003c/em\u003e, free/libre and open-source software) platform [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e] collaborative, interoperable mapping platform in which we instantiated our thematic atlases with choropleth maps and their legends. Our mortality atlas showcase represents good potential for attractiveness and use, thanks to its strengths in terms of - display and visualization of results - ergonomics (usability) and in terms of - understanding and semantic interpretation of cartographic and statistical results. To achieve the desired atlas results, the uMap platform did not require any additional IT development or management of the remote uMap server (handled by \u003cem\u003eOSM.fr\u003c/em\u003e). The standards described in the \u003cspan refid=\"Sec13\" class=\"InternalRef\"\u003ematerial\u003c/span\u003e section are used in our work.\u003c/p\u003e \u003cp\u003eThe \"\u003cem\u003eFeuTricolore\u003c/em\u003e\" (TrafficLight) atlas for France appears to be highly relevant in terms of the instant message it communicates, and in this respect, we have provided a complement to the existing covid 19 pandemic cartography that is easy to interpret for any audience.\u003c/p\u003e \u003cp\u003eTo join the collaborative platform concept, our mortality atlases will be open to sharing since mortality data sourced by \u003cem\u003eINSEE\u003c/em\u003e is open and not subject to health data regulations like the \u003cem\u003eR\u0026egrave;glement G\u0026eacute;n\u0026eacute;ral sur la Protection des Donn\u0026eacute;es (RGPD/GDPR\u003c/em\u003e, general data protection regulation). We expect quantitative and critical feedback from users of the uMap environment. Also, the map search by keyword, will enable everyone to access our mortality atlases.\u003c/p\u003e \u003cp\u003eIn terms of geomatics, since the two-step action plan \u0026ldquo;without and with webmapping\u0026rdquo; is rather complex, it would be advisable to consider a number of improvements (reduce the number of software packages used, plan interconnections between software packages to automate processing chains, give priority to geomatic processing in step 1, if possible, only in the R environment). And for step 2, development should be planned with the help of uMap contributors to facilitate automation of the creation of double legends for choropleth maps (in color ranges).\u003c/p\u003e \u003cp\u003eAt the epidemiological level, our mortality atlases can prove extremely useful for health crisis management if they are developed in real time in uMap or, as in our case, in a retrospective framework on a collaborative platform. Finally, to improve our understanding of the municipal landscape, following the detection of alerts of excess- or under mortality, we could consider coupling our results with socio-demographic data from the municipalities.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to thank the creators and contributors to the uMap platform, especially Yohan BONIFACE, as well as the geomatics world in particular: the OSGeo and OSGeo.fr geospatial open-source foundations and the Open Street Map association. We would also like to thank all \u003cem\u003eINSEE\u003c/em\u003e\u0026apos;s open data suppliers, notably data-gouv, as well as all the developers who make their software available free of charge. Last but not least, we would like to extend our special thanks to our healthcare establishments, who through our activities contribute to public health research and the development of geomatics in healthcare. The authors would like to thank Frank DUFOUR, of the Facult\u0026eacute; de M\u0026eacute;decine de Nice, RETINES laboratory, for his help in revising the English version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTranslation and publication costs are funded by the \u003cem\u003eGroupement des H\u0026ocirc;pitaux de l\u0026apos;Institut Catholique de Lille\u0026nbsp;\u003c/em\u003e\u003cem\u003e(GHICL\u003c/em\u003e, Group of hospitals of the Institut Catholique de Lille). Ms. Quesnel\u0026apos;s research activity is supported by the Lille University Hospital. The RETINES laboratory has made available bibliographic referencing software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe gross \u003cem\u003eINSEE\u003c/em\u003e death dataset extracted May 22, 2020 supporting the conclusions of this article is available in the following healthDataset repository , in [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors details\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003eLille University Hospital (CHU de Lille), Regional House of Clinical Research \u003cstrong\u003e(MRRC),\u0026nbsp;\u003c/strong\u003ePublic Health, Lille, France.\u0026nbsp;\u003cstrong\u003eAnne QUESNEL-BARBET (AQB)\u003c/strong\u003e, [email protected]; [email protected], https://orcid.org/0000-0003-1038-7344\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e2\u003c/sup\u003eTITSOFT, Montpellier, France. \u003cstrong\u003eThierry PAGES (TP)\u003c/strong\u003e, [email protected]; https://orcid.org/0000-0002-6075-7633\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e3\u003c/sup\u003eCERIM Lab - EA 2694: Public Health, Faculty of Medicine, University of Lille, Lille, Nord, France.\u0026nbsp;\u003cstrong\u003eJulien SOULA (JS)\u0026nbsp;\u003c/strong\[email protected]; https://orcid.org/0000-0003-0875-7713\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e4\u003c/sup\u003eRETINES Laboratory - Faculty of Medicine, University of C\u0026ocirc;te d\u0026apos;Azur (UCA), Nice, France, \u003cstrong\u003eGilles MAIGNANT (GM)\u003c/strong\u003e, [email protected] ; https://orcid.org/0000-0003-2017-3972\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e5\u003c/sup\u003eDSI/CIO/CMIO - Catholic Hospitals of Lille (GHICL), Nord, Lille, France. \u003cstrong\u003eArnaud HANSSKE (AH)\u003c/strong\u003e,\u0026nbsp;\u003ca href=\"mailto:[email protected]\"\[email protected]\u0026nbsp;\u003c/a\u003e; https://orcid.org/0000-0001-7029-2318\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorresponding author\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnne QUESNEL-BARBET,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eE-mail address: [email protected]; [email protected]\u003c/p\u003e\n\u003cp\u003eCHU de Lille, Maison R\u0026eacute;gionale de la Recherche Clinique, Sant\u0026eacute; Publique, Lille, France.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy design: AQB.\u003c/p\u003e\n\u003cp\u003eData preparation: TP, AQB.\u003c/p\u003e\n\u003cp\u003eServer management: JS\u003c/p\u003e\n\u003cp\u003eStudy and statistical analysis: AQB, TP, GM.\u003c/p\u003e\n\u003cp\u003eDrafting and finalization of manuscript: AQB, GM, TP, AH.\u003c/p\u003e\n\u003cp\u003eAll authors have read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis article does not contain any studies with human participants or animals performed by any of the authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatement on conflicts of interest\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eFry R, Hollinghurst J, Stagg HR, Thompson DA, Fronterre C, Orton C, Lyons RA, Ford DV, Sheikh A, Diggle PJ: \u003cstrong\u003eReal-time spatial health surveillance: Mapping the UK COVID-19 epidemic\u003c/strong\u003e. \u003cem\u003eInt J Med Inform\u0026nbsp;\u003c/em\u003e2021, \u003cstrong\u003e149\u0026nbsp;\u003c/strong\u003e(104400):1-8. https://doi.org/10.1016/j.ijmedinf.2021.104400.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWeb mapping or online mapping.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Web_mapping]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eBrunet R: \u003cstrong\u003eMap and sketch: report on Arnould C, Denisart F, Mangin C, 1999, \u0026quot;Geography sketches for the baccalaur\u0026eacute;at\u0026quot;, Paris: Ellipses. And on: Poidevin D, 1999, \u0026quot;The Action Map. A practical guide to map design and production\u0026quot;, Paris: Ellipses.\u003c/strong\u003e \u003cem\u003eMappemonde\u0026nbsp;\u003c/em\u003e2000, \u003cstrong\u003e60\u003c/strong\u003e(4):45-46.\u003c/li\u003e\n \u003cli\u003ePoidevin D: \u003cstrong\u003eThe Map as a Means of Action.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eA practical guide to card design and production / \u003cem\u003eLa Carte moyen d\u0026rsquo;action. Guide pratique pour la conception et la r\u0026eacute;alisation de cartes\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u0026nbsp;\u003c/em\u003eParis, France: Ellipses; 1999.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eRavenea\u003c/em\u003eu J: \u003cstrong\u003eBrunet R, (1987) The map, instructions for use / \u003cem\u003eLa carte, mode d\u0026rsquo;emploi.\u003c/em\u003e Paris, Fayard/Reclus, 270 p., 269 ill.\u003c/strong\u003e \u003cem\u003eCahiers de g\u0026eacute;ographie du Qu\u0026eacute;bec\u0026nbsp;\u003c/em\u003e1988, \u003cstrong\u003e32\u0026nbsp;\u003c/strong\u003e(85):99\u0026ndash;102. https://doi.org/10.7202/021949ar.\u003c/li\u003e\n \u003cli\u003eQuesnel-Barbet A, Dufossez F, Souris M, Delerue D, Hansske A: \u003cstrong\u003eGeomatics in health and applications in the north of France\u003c/strong\u003e. \u003cem\u003eTechniques de l\u0026apos;ing\u0026eacute;nieur: \u0026Eacute;lectronique, robotique et TIC pour la sant\u0026eacute;\u0026nbsp;\u003c/em\u003e2016, \u003cstrong\u003eTIB 628\u0026nbsp;\u003c/strong\u003e(med4000):5; 1-41. https://doi.org/10.51257/a-v1-med4000.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eThe Home of Location Technology Innovation and Collaboration.\u0026nbsp;\u003c/strong\u003e[https://www.ogc.org/standards/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eFAIR data - FAIR principles, as published by GO FAIR - Findable, Accessible, Interoperable and Reusable (FAIR).\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/FAIR_data]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eHayat. Flora: \u003cstrong\u003eProduction of digital commons and cartographic use\u003c/strong\u003e. Paris, France: Paris Cit\u0026eacute;; 2019.[http://www.theses.fr/2019UNIP7135/document]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWelcome to the OpenStreetMap wiki.\u0026nbsp;\u003c/strong\u003e[https://wiki.openstreetmap.org/wiki/Main_Page]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOpenStreetMap (OSM)\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/OpenStreetMap]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOpen data: the opening up of public data - Free software issues, open standards and interoperability / L\u0026apos;open data: l\u0026apos;ouverture des donn\u0026eacute;es publiques - \u003cem\u003eEnjeux du logiciel libre, standards ouverts et interop\u0026eacute;rabilit\u0026eacute;.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[http://www.apitux.org/index.php?2010/11/19/235-opendata-l-ouverture-des-donnees-publiques]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eBECQUET J-C: \u003cstrong\u003eOpen GIS, software and data conference: technical training for students, professionals and jobseekers / \u003cem\u003eColloque SIG libres, logiciels et donn\u0026eacute;es: des formations techniques pour \u0026eacute;tudiants, professionnels et demandeurs d\u0026apos;emploi\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e In: \u003cem\u003eLe monde du libre: le logiciel libre et les donn\u0026eacute;es ouvertes au service des territoires: 30 juin 2006; XSALTO, Digne-les-Bains, France.\u003c/em\u003e: APITUX; 2006. [https://www.agendadulibre.org/events/587; http://www.apitux.org/index.php?q=COLLOQUE]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCreate interactive and crowd-sourced web maps.\u0026nbsp;\u003c/strong\u003e[https://about.canvis.app/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCrowdsourcing - Participatory production - Open outsourcing.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Crowdsourcing]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCrowdmapping.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Crowdmapping]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eNoucher M, Gautreau P: \u003cstrong\u003eIs open access reshuffling the deck?\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eNew perspectives for geographical data / \u003cem\u003eLe libre acc\u0026egrave;s rebat-il les cartes? Nouvelles perspectives pour les donn\u0026eacute;es g\u0026eacute;ographiques.\u003c/em\u003e\u003c/strong\u003e \u003cem\u003eLes Cahiers du num\u0026eacute;rique\u0026nbsp;\u003c/em\u003e2013, \u003cstrong\u003e9\u003c/strong\u003e(1):57-83.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOpen source.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Open_source]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWeb 2.0, sometimes called participative Web.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Web_2.0]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eBECQUET J-C: \u003cstrong\u003eOpen formats and interoperability / \u003cem\u003eFormats ouverts et interop\u0026eacute;rabilit\u0026eacute;\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e2004, \u003cstrong\u003e84\u003c/strong\u003e(84).[http://www.apitux.org/index.php?2008/07/25/84-formats-ouverts-et-interoperabilite]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eInteroperability.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Interoperability]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWhat is a data pipeline?\u0026nbsp;\u003c/strong\u003e[https://www.ibm.com/topics/data-pipeline]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWhat is a data pipeline and how does it work?\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eDefinition of a data pipeline / \u003cem\u003eQu\u0026rsquo;est-ce qu\u0026rsquo;un data pipeline et comment cela fonctionne? D\u0026eacute;finition d\u0026apos;un pipeline de donn\u0026eacute;es.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.talend.com/fr/resources/what-is-a-data-pipeline/#:~:text=Un%20pipeline%20de%20donn%C3%A9es%20englobe,donn%C3%A9es%20pr%C3%AAtes%20%C3%A0%20%C3%AAtre%20exploit%C3%A9es]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eBERNARD M, BRIONE M: \u003cstrong\u003eGIS Observatory: Norms and standards, quo vadis? / \u003cem\u003eObservatoire du SIG: Normes et standards, quo vadis?\u003c/em\u003e\u003c/strong\u003e In: \u003cem\u003eObservatoire du SIG.\u003c/em\u003e Edited by CNIG. Paris, France: SIGMAG TV.fr; 2022: 1-10. [https://cnig.gouv.fr/IMG/pdf/_c_vpw_sigmag_sigtv.fr_n34_obervatoire_du_sig_enquete_normes_standards_-_octobre_2022.pdf]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eFrench Association for Standardization (AFNOR.org) in cooperation with the International Organization for Standardization (ISO.org)\u0026nbsp;\u003c/strong\u003e[https://www.iso.org/organization/1738.html]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOpen Geospatial Consortium (OGC), Inc.\u0026nbsp;\u003c/strong\u003e[https://www.iso.org/organization/9825.html]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eEtalab: a French public administration to improve public service and public action through data / \u003cem\u003eEtalab: une administration fran\u0026ccedil;aise pour am\u0026eacute;liorer le service public et l\u0026apos;action publique gr\u0026acirc;ce aux donn\u0026eacute;es.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://fr.wikipedia.org/wiki/Etalab]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eData Act: Proposal For A Regulation On Harmonised Rules On Fair Access To And Use Of Data.\u0026nbsp;\u003c/strong\u003e[https://digital-strategy.ec.europa.eu/en/library/data-act-proposal-regulation-harmonised-rules-fair-access-and-use-data]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGeodetic parameter data set EPSG.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/EPSG_Geodetic_Parameter_Dataset#:~:text=EPSG%3A4326%20%2D%20WGS%2084%2C,including%20Google%20Maps%20and%20OpenStreetMap]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eISO 19111:2019 Geographic information \u0026mdash; Coordinate reference system.\u0026nbsp;\u003c/strong\u003e[https://www.iso.org/standard/74039.html]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eEPSG:4326 -- WGS84 - World Geodetic System 1984, used in GPS.\u0026nbsp;\u003c/strong\u003e[https://epsg.io/4326]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eIT applications in science; Including digital geographic information.\u0026nbsp;\u003c/strong\u003e[https://www.iso.org/ics/35.240.70/x/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDeveloping With Web Standards \u0026ndash; Recommendations and best practices.\u0026nbsp;\u003c/strong\u003e[https://www.456bereastreet.com/lab/developing_with_web_standards/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eSarafinof D, Mistre A, Picinbono G, Vallet B, Heydel L: \u003cstrong\u003eThe GeoBIM approach: from land management to building management / \u003cem\u003eLa d\u0026eacute;marche G\u0026eacute;oBIM: de la gestion du territoire \u0026agrave; celle d\u0026rsquo;un b\u0026acirc;timent\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e \u003cem\u003eAnnales des Mines - Responsabilit\u0026eacute; et environnement\u0026nbsp;\u003c/em\u003e2019, \u003cstrong\u003e94\u003c/strong\u003e(2):42-46. https://doi.org/10.3917/re1.094.0042.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCityGML GeoBIM ADE - This GeoBIM standard was launched in 2009\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e: [https://www.citygmlwiki.org/index.php?title=Citygml_Wiki]; 2009. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDGIWG: The Defence Geospatial Information Working Group.\u0026nbsp;\u003c/strong\u003e[https://dgiwg.org/about]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eScancovid.\u0026nbsp;\u003c/strong\u003e[https://scancovid.scansante.fr/scancovid/_w_70c5f626/?code=open_data]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCare and territories / Soins-et-territoires.\u0026nbsp;\u003c/strong\u003e[https://www.scansante.fr/actualites/soins-et-territoires]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eCromley EK, McLafferty SL: \u003cstrong\u003eGIS and Public Health\u003c/strong\u003e: Guilford Press; 2011.\u003c/li\u003e\n \u003cli\u003eJones K, Duncan C, Moon G: \u003cstrong\u003eContext, scale, and the geography of health: understanding the relevance of ecological studies\u003c/strong\u003e. \u003cem\u003eAmerican Journal of Public Health\u0026nbsp;\u003c/em\u003e1998, \u003cstrong\u003e88\u003c/strong\u003e(2):216-222.\u003c/li\u003e\n \u003cli\u003eKwan MP: \u003cstrong\u003eThe uncertain geographic context problem\u003c/strong\u003e. \u003cem\u003eAnnals of the Association of American Geographers\u0026nbsp;\u003c/em\u003e2013, \u003cstrong\u003e103\u003c/strong\u003e(2):221-244. https://doi.org/10.1080/00045608.2012.687349.\u003c/li\u003e\n \u003cli\u003eKazi AM, Ali M, Ayub K, Kalimuddin H, Zubair K, Kazi AN, Artani A, Ali SA: \u003cstrong\u003eGeo-spatial reporting for monitoring of household immunization coverage through mobile phones: Findings from a feasibility study\u003c/strong\u003e. \u003cem\u003eInt J Med Inform\u0026nbsp;\u003c/em\u003e2017, \u003cstrong\u003e107\u003c/strong\u003e:48-55. https://doi.org/10.1016/j.ijmedinf.2017.09.004.\u003c/li\u003e\n \u003cli\u003ePlantier M, Havet N, Durand T, Caquot N, Amaz C, Philip I, Biron P, Perrier L: \u003cstrong\u003eDoes adoption of electronic health records improve organizational performances of hospital surgical units? Results from the French e-SI (PREPS-SIPS) study\u003c/strong\u003e. \u003cem\u003eInt J Med Inform\u0026nbsp;\u003c/em\u003e2017, \u003cstrong\u003e98\u003c/strong\u003e:47-55. https://doi.org/10.1016/j.ijmedinf.2016.12.002.\u003c/li\u003e\n \u003cli\u003eAlam K, Mahumud RA, Alam F, Keramat SA, Erdiaw-Kwasie MO, Sarker AR: \u003cstrong\u003eDeterminants of access to eHealth services in regional Australia\u003c/strong\u003e. \u003cem\u003eInt J Med Inform\u0026nbsp;\u003c/em\u003e2019, \u003cstrong\u003e131\u003c/strong\u003e:103960. https://doi.org/10.1016/j.ijmedinf.2019.103960.\u003c/li\u003e\n \u003cli\u003eNaghavi M, Mohsen N, Contributors.: \u003cstrong\u003eGlobal, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013\u003c/strong\u003e. \u003cem\u003eLancet\u0026nbsp;\u003c/em\u003e2015, \u003cstrong\u003e385\u003c/strong\u003e(9963):117-171. https://doi.org/10.1016/s0140-6736(14)61682-2.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCancers in France 2015 edition / \u003cem\u003eLes cancers en France \u0026eacute;dition 2015.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.e-cancer.fr/content/download/148692/1867381/file/Les-cancers-en-France-edition-2015.pdf]. Accessed 17 july 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGeodes: Geo Data in Public Health, the cartographic observatory of epidemiological indicators / \u003cem\u003eGeodes: G\u0026eacute;o Donn\u0026eacute;es En Sant\u0026eacute; Publique, l\u0026rsquo;observatoire cartographique des indicateurs \u0026eacute;pid\u0026eacute;miologiques.\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e[https://geodes.santepubliquefrance.fr/#view=map2\u0026amp;c=indicator]. Accessed 17 July 2024..\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eThe health monitoring institute / \u003cem\u003eL\u0026apos;institut de veille sanitaire.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.has-sante.fr/jcms/c_436647/fr/institut-de-veille-sanitaire#:~:text=L\u0026apos;Institut%20de%20veille%20sanitaire,la%20population%20et%20son%20%C3%A9volution.]. Accessed 17 July 2024..\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSentinelles.\u0026nbsp;\u003c/strong\u003e[https://www.sentiweb.fr/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSurvey: Coronavirus: which towns in the Nord and Pas-de-Calais departments have recorded higher mortality since 1 March? / \u003cem\u003eEnqu\u0026ecirc;te: Coronavirus : quelles communes du Nord et du Pas-de-Calais enregistrent une surmortalit\u0026eacute; depuis le 1er mars ?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://france3-regions.francetvinfo.fr/hauts-de-france/enquete-coronavirus-quelles-communes-du-nord-du-pas-calais-enregistrent-surmortalite-1er-mars-1825674.html?fbclid=IwAR0ue9PUte3GmDvyRKI44WHl2An4YTjrbHUl8pFTiSEOmg7gz3BvrkTuQ-M#La%20surmortalit%C3%A9%20dans%20l\u0026apos;arrondissement%20de%20Lille]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eWHO.: \u003cstrong\u003eWorld health statistics 2014\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by World-Health-Organization: Apps. [https://www.who.int/news/item/15-05-2014-world-health-statistics-2014]; 2014. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eQuesnel-Barbet A, Soula J, Sauleau E-A, Bazile P, Dufossez F, Albert F, Hansske A, Trilling L, Maignant G: \u003cstrong\u003ePoleSat, health planning modelling for decision support. Application to a reorganization of vascular catheterization provision, at the geographic level of \u0026quot;territorial hospital grouping\u003c/strong\u003e. \u003cem\u003eISTE openscience\u0026nbsp;\u003c/em\u003eOct 2022, \u003cstrong\u003eNum\u0026eacute;ro sp\u0026eacute;cial: Gestion et ing\u0026eacute;nierie des syst\u0026egrave;mes hospitaliers\u0026nbsp;\u003c/strong\u003e(5):21. https://doi.org/10.21494/ISTE.OP.2022.0883.\u003c/li\u003e\n \u003cli\u003eQuesnel-Barbet A, Soula J, Sauleau E-A, Bazile P, Dufossez F, Maignant G, Albert F, Hansske A, Trilling L: \u003cstrong\u003ePoleSat, health planning modelling for decision support. Application to a reorganization of vascular catheterization provision, at the geographic level of \u0026quot;territorial hospital grouping\u0026nbsp;\u003c/strong\u003e\u003cem\u003eISTE openscience\u0026nbsp;\u003c/em\u003eJanv 2022, \u003cstrong\u003eNum\u0026eacute;ro sp\u0026eacute;cial: Gestion et ing\u0026eacute;nierie des syst\u0026egrave;mes hospitaliers\u0026nbsp;\u003c/strong\u003e(5):27. https://doi.org/10.21494/ISTE.OP.2022.0774.\u003c/li\u003e\n \u003cli\u003eQuesnel A, Soula J, Wyndels K, Beuscart R: \u003cstrong\u003ePOLESAT: e-Atlas of Health, Planning and Sustainable Development\u003c/strong\u003e. \u003cem\u003eIng\u0026eacute;nierie et recherche biom\u0026eacute;dicale\u0026nbsp;\u003c/em\u003e2009, \u003cstrong\u003e30\u003c/strong\u003e(4):201-204. https://doi.org/10.1016/j.irbm.2009.05.005.\u003c/li\u003e\n \u003cli\u003eQuesnel-Barbet A, Soula J, Dufossez F, Sharma N, Ruhela A, Beuscart R: \u003cstrong\u003ePOLESAT, an innovative e-health geomatic platform in decision-making based on: Geographical approach, medical knowledge visualization and geographic information system and web-mapping architecture\u003c/strong\u003e. \u003cem\u003eIRBM\u0026nbsp;\u003c/em\u003e2013, \u003cstrong\u003e34\u003c/strong\u003e(4-5):267-273. https://doi.org/10.1016/j.irbm.2013.06.001.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDomain name.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Domain_name]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eReview 2018 OSM France services /\u003cem\u003eRevue 2018 Les services OSM France.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.openstreetmap.fr/revue-2018-les-services-osm-france/]. Accessed 27 June 2023.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eFrench all-cause mortality monitoring atlas (health crisis - SARS COV-2). Period from 1st March to 15th May inclusive 2020 - 1st containment. Excess mortality alerts expressed by Traffic-Light - uMap.\u0026nbsp;\u003c/strong\u003e[http://umap.openstreetmap.fr/en/map/polesatatlasmortalite2020_feutricolore_v1_580435#6/46.793/4.570]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eShahamati S, Baumann Y, Caquard S, Denieul-Pinsky L, Shaw E: \u003cstrong\u003euMap: A Free, Open-Source Alternative to Google My Maps\u003c/strong\u003e. \u003cem\u003eCartographicPerspectivesorg\u0026nbsp;\u003c/em\u003e2022(99):6\u0026ndash;18. https://doi.org/10.14714/CP99.1729.\u003c/li\u003e\n \u003cli\u003eGentleman R, Ihaka R, R Core Team., Bates D, Chambers J, Dalgaard P, Gentilhomme R, Hornik K, Isaac R, Kalibera T\u003cem\u003e\u0026nbsp;et al\u003c/em\u003e: \u003cstrong\u003eThe R Project. The Comprehensive R Archive Network: R for Windows\u003c/strong\u003e. V 4.3.1 edn Global community; 1997: GNU GPL v2. [cran.r-project]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eQGIS-Development-Team.: \u003cstrong\u003eQGIS is A Free, Open Source, Cross Platform (lin/win/mac) Geographic Information System (GIS)\u003c/strong\u003e. V 3.18.1-Z\u0026uuml;rich edn. Global community: Qgis.org; 2002: CC BY-SA 3.0; GPL-2.0. [github.com/qgis/QGIS]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDigital cartography workshop with uMap / \u003cem\u003eAtelier cartographie num\u0026eacute;rique avec uMap\u003c/em\u003e.\u0026nbsp;\u003c/strong\u003e[https://associationici.fr/wp-content/uploads/2018/09/Guide-de-prise-en-main-Umap-OSM-v3.pdf]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGross\u003c/strong\u003e \u003cstrong\u003edeath file from 2017 to 2020: Daily death files download site - Latest files released - Daily death files by sex, age and place - May 22, 2020 / \u003cem\u003eFichier de d\u0026eacute;c\u0026egrave;s brut de 2017 \u0026agrave; 2020: Site de t\u0026eacute;l\u0026eacute;chargement des fichiers des d\u0026eacute;c\u0026egrave;s quotidiens - Derniers fichiers diffus\u0026eacute;s - Fichiers des d\u0026eacute;c\u0026egrave;s quotidiens par sexe, \u0026acirc;ge et lieu - 22 mai 2020\u003c/em\u003e.\u0026nbsp;\u003c/strong\u003e[https://www.insee.fr]; [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/Gross_death_file_insee_extraction_20200522.zip]; https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/deathDataset]; [https://www.data.gouv.fr/fr/datasets/fichier-des-personnes-decedees/#/resources]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCoronavirus: why the errors in counting deaths in nursing homes? / \u003cem\u003eCoronavirus: pourquoi ces erreurs dans les d\u0026eacute;comptes des d\u0026eacute;c\u0026egrave;s en Ehpad?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.leparisien.fr/societe/coronavirus-pourquoi-ces-erreurs-dans-les-decomptes-des-deces-en-ehpad-20-05-2020-8320674.php]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSARS-CoV-2, 68 communes have reached the alert threshold - (alert threshold set at 20 new contaminations per week per 100,000 inhabitants) / \u003cem\u003eCovid-19 : en Belgique, 68 communes ont atteint le seuil d\u0026apos;alerte -( seuil d\u0026apos;alerte fix\u0026eacute; \u0026agrave; 20 nouvelles contaminations par semaine pour 100 000 habitants\u003c/em\u003e).\u0026nbsp;\u003c/strong\u003e[https://france3-regions.francetvinfo.fr/hauts-de-france/covid-19-belgique-soixante-communes-ont-atteint-seuil-alerte-1856270.html?fbclid=IwAR2eSX_aVs1rPn7qLU_ZiNKPos0_XVl_q4IRmAy5seknZaLFu0AHn93Q2fE#xtor=EPR-521-[france3regions]-20200723-[info-titre1]\u0026amp;pid=726375-1583524212-a50ad1ea]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCovid-19: what the 150 French clusters reveal about the epidemic / \u003cem\u003eCovid-19: ce que r\u0026eacute;v\u0026egrave;lent les 150 clusters fran\u0026ccedil;ais de l\u0026rsquo;\u0026eacute;pid\u0026eacute;mie.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.lefigaro.fr/sciences/covid-19-ce-que-revelent-les-150-clusters-francais-de-l-epidemie-20200605?origine=VWT16001\u0026amp;utm_campaign=echobox\u0026amp;utm_medium=Social\u0026amp;utm_source=Facebook\u0026amp;fbclid=IwAR2_RHpbQK7c1Urqlw-TTQJjJ_EuhCKnW3M-X6na6rgQKxt6rheqfJ22NLU#Echobox=1591387766]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDownload the 2017 legal population file (excluding Mayotte) / \u003cem\u003eT\u0026eacute;l\u0026eacute;chargement du fichier d\u0026apos;ensemble des populations l\u0026eacute;gales en 2017 (hors Mayotte).\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.insee.fr/fr/statistiques/4265429]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eLegal population of Mayotte in 2017, Population census / \u003cem\u003ePopulations l\u0026eacute;gales de Mayotte en 2017 - Recensement de la Population.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.insee.fr/fr/statistiques/2120838]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eADMIN-EXPRESS-COG 2020 edition by territory: the administrative division of the French territory (commune, departmental district, department, region...) / \u003cem\u003eADMIN-EXPRESS-COG 2020 \u0026eacute;dition par territoire: la division administrative du territoire fran\u0026ccedil;ais (commune, arrondissement, d\u0026eacute;partement, r\u0026eacute;gion...).\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://geoservices.ign.fr/adminexpress]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003ePages T, Soula J, Quesnel-Barbet A: \u003cstrong\u003eDeath dataset from 2018-2020 linked to supplementary_file_1 / \u003cem\u003eJeu de donn\u0026eacute;es sur les d\u0026eacute;c\u0026egrave;s de 2018-2020 li\u0026eacute;s au fichier suppl\u0026eacute;mentaire_1\u003c/em\u003e\u003c/strong\u003e. [https://thymine.univ-lille.fr/PoleSat_mortality_atlas/deathDataset/SF1_supplementary_file_1.zip]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eWagniez P: \u003cstrong\u003ePhilcarto utility guide: Eclats2.xxDOC / \u003cem\u003eGuide de l\u0026apos;utilitaire pour Philcarto: Eclats2.xxDOC\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e 2019 [http://philcarto.free.fr/02_telecharger_logiciels/02_3_telecharger_eclats/fichiers_installer_eclats/Eclats2.xxDOC.zip]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eWagniez P, Contributors.: \u003cstrong\u003ePhilcarto: Eclat utility / \u003cem\u003ePhilcarto: utilitaire Eclat\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e V 2.10.h. France: 2023: freeware. [http://philcarto.free.fr/02_telecharger_logiciels_nouveaux_2023/02_1_telecharger_2023.html]; Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003edata.gouv.fr., Lexman A: \u003cstrong\u003eMap of departments, Metropolitan France and DROM - Open Data Commons Open Database License (ODbL) / \u003cem\u003eCarte des d\u0026eacute;partements, France M\u0026eacute;tropolitaine et DROM - ODbL\u003c/em\u003e\u003c/strong\u003e. [https://www.data.gouv.fr/fr/datasets/carte-des-departements-2-1/; https://github.com/gregoiredavid/france-geojson/blob/master/departements.geojson]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003edata.gouv.fr., Lexman A: \u003cstrong\u003eMap of French regions (new regions in 2016) / \u003cem\u003eCarte des r\u0026eacute;gions fran\u0026ccedil;aises (nouvelles r\u0026eacute;gions de 2016)\u003c/em\u003e\u003c/strong\u003e. [https://www.data.gouv.fr/fr/datasets/carte-des-regions-francaises-nouvelles-regions-de-2016/;\u003c/li\u003e\n \u003cli\u003ehttps://portailsig.org/content/recuperer-des-donnees-openstreetmap-gdalogr.html]. Accessed 16 July 2024. data.gouv.fr., OSM-Contributors.: \u003cstrong\u003eLocation of intensive care units - services-de-reanimation.geojson Open Data Commons Open Database License (ODbL) / \u003cem\u003eLocalisation des services hospitaliers de r\u0026eacute;animation - services-de-reanimation.geojson, (ODbL)\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e [https://www.data.gouv.fr/fr/datasets/localisation-des-services-hospitaliers-de-reanimation/]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003edata.gouv.fr., OSM-Contributors.: \u003cstrong\u003eLocation of emergency services: data - services-daccueil-des-urgences.geojson - Open Data Commons Open Database License (ODbL) / \u003cem\u003eLocalisation des services d\u0026apos;accueil des urgences: data - services-daccueil-des-urgences.geojson (ODbL)\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u003c/em\u003e [https://www.data.gouv.fr/fr/datasets/localisation-des-services-daccueil-des-urgences/#/resources]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003egeo.data.gouv.fr: \u003cstrong\u003eSwitching off geo.data.gouv.fr datasets- France - Emergency medicine and authorizations / \u003cem\u003eExtinction des jeux de donn\u0026eacute;es sur geo.data.gouv.fr - France - M\u0026eacute;decine d\u0026apos;urgence et autorisations\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e.\u0026nbsp;\u003c/em\u003e[https://www.data.gouv.fr/fr/posts/extinction-de-geo-data-gouv-fr/]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSQLite: A Database Engine Multiplatform - Public Domain License.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/SQLite]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003eHipp D-R: \u003cstrong\u003eSQLite: Most Widely Deployed and Used Database Engine\u003c/strong\u003e. V3.42.0 edn. North Carolina, USA; 2000: Open-Source, not Open-Contribution. [sqlite.org]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eHelp - updating R, Rstudio and packages / \u003cem\u003eAide - mise \u0026agrave; jour de R, Rstudio et packages.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://delladata.fr/tutoriel-mise-a-jour-de-r/]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003eAllaire JJ, Gutman J, Mcphers J, Cheng J, De-Vries A, Chen D, Arendt C, Scheidegger C, Falbel D, Aden-Buie G\u003cem\u003e\u0026nbsp;et al\u003c/em\u003e: \u003cstrong\u003eRStudio IDE\u003c/strong\u003e. V 2023.06.1+524 edn. Global community, 2011: Open Source (AGPL v3). [posit.co/products/open-source/rstudio/]; [github.com/rstudio/rstudio]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eWickham H-A: \u003cstrong\u003ePlyr library: Tools for Splitting, Applying and Combining Data\u003c/strong\u003e. V1.8.7 edn. Houston, TX, USA: 2014: Copyright Holder: Hadley Wickham, MIT license. [cran.r-project]; [plyr.had.co.nz]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eAllaire J, Xie Y, Dervieux C, McPherson J, Luraschi J, Ushey K, Atkins A, Wickham H, Cheng J, Chang W\u003cem\u003e\u0026nbsp;et al\u003c/em\u003e: \u003cstrong\u003eRmarkdown: Dynamic Documents for R. R package\u003c/strong\u003e. V 2.21 edn. Global community: cran.r-project.org; 2007: GPL-3. [github.com/rstudio/rmarkdown]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eWalker A, Schauberger P, Braglia L, Sturm J, Garbuszus J-M, Barbone J-M, Contributors.: \u003cstrong\u003eOpenxlsx library: Read, Write and Edit xlsx Files\u003c/strong\u003e. V 4.2.5 edn. Sidney, Australia; Vienne, Austria 2014: MIT-license. [https://ycphs.github.io/openxlsx/]; [github.com/ycphs/openxlsx]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGDAL: an MIT-licensed open-source translator library for raster and vector geospatial data formats - the May 2023 GDAL/OGR v3.7.0.\u0026nbsp;\u003c/strong\u003e[https://github.com/OSGeo/]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003eWANIEZ P: \u003cstrong\u003ePhilcarto\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e, v2021.d edn. Bordeaux, France; 2010: freeware. [Philcarto.free.fr]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eBloch M: \u003cstrong\u003eMapshaper: Tools for editing Shapefile, GeoJSON, TopoJSON and CSV files\u003c/strong\u003e. V0.6.33 edn. NY, USA: 2014: MPL 2.0 license. [mapshaper.org]; [github.com/mbloch/mapshaper/wiki]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eMacWright T, Whong C, Contributors.: \u003cstrong\u003eGEOJSON file editor: geojson.io\u003c/strong\u003e. In: \u003cem\u003eOSM-mapboxcom - Maxarcom.\u003c/em\u003e Edited by rolling (2014-08-11) edn. NY, USA; Global community 2022: ISC License. [geojson.io] ; [github.com/mapbox/geojson.io]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eGougelet P-E, Mezich T, Budz45.: \u003cstrong\u003eXnView is a general-purpose image organizer and file manager.\u0026nbsp;\u003c/strong\u003eProprietary license, V 2.51.2 edn. Reims, France; 1998, [xnview.com/fr/about/]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eGougelet P-E: \u003cstrong\u003eXnConvert.\u0026nbsp;\u003c/strong\u003eV 1.98 - Windows x64; v 7.99 Libformat edn. Reims, France: Dec 2021: Proprietary license. [newsgroup.xnview.com/viewforum.php?f=79]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eIrfan S: \u003cstrong\u003eIrfan\u0026nbsp;\u003c/strong\u003ein: \u003cem\u003eSupports Windows XP, Vista, 7, 8, 10 and 11.\u003c/em\u003e V 4.5.9 edn. Austria; 1996-2023: Freeware for non-commercial use. [irfanview.com]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eAdobe.: \u003cstrong\u003ePhotoshop\u003c/strong\u003e. V 7.0 edn. Mountain View, California, USA: 1990-2002: Proprietary license. [Adobe.com]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003eMicrosoft: \u003cstrong\u003ePaint.\u0026nbsp;\u003c/strong\u003eV 11.2301.22.0 edn. USA: 2000: Proprietary license. [support.microsoft.com]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003edotPDN-LLC., Brewster R, Contributors.: \u003cstrong\u003ePaint.NET image and photo editing software for Windows PCs.\u003c/strong\u003e V 5.0.7 edn. Washington, USA; 2019: Classic, free-of-charge. [Getpaint.net]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eVisual Basic for Applications.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Visual_Basic_for_Applications]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eTan V: \u003cstrong\u003eOpen-source developer-friendly spreadsheet library compatible with Microsoft Excel\u003c/strong\u003e. V 3.5.0 (Sept 2020) edn. Singapore: 2011: MIT License. [spreadsheetlight.com]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eBoniface Y: \u003cstrong\u003euMap-project and source code\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by github.com, v 1.4.0 edn. France: 2013: WTFPL license. [github.com/umap-project/umap/]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDo What The Fuck You Want To Public License (WTFPL).\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/WTFPL]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eBoniface Y: \u003cstrong\u003euMap\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by github.com, v 1.4.0 edn. France; 2013: WTFPL license. [https://umap.openstreetmap.fr/en/]; [https://github.com/umap-project/umap/issues/1115]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003euMap.\u0026nbsp;\u003c/strong\u003e[https://wiki.openstreetmap.org/wiki/UMap]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eWhat\u0026apos;s the Difference Between Domain Names and Web Hosting? / \u003cem\u003eQuelle Diff\u0026eacute;rence Entre Nom de Domaine et H\u0026eacute;bergement Web?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.hostinger.fr/tutoriels/nom-de-domaine-et-hebergement#Quelle_est_la_difference_entre_nom_de_domaine_et_hebergement_web]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003eGoogle.: \u003cstrong\u003eGoogle Earth For 3D representation of Earth\u003c/strong\u003e. V 7.3.6.9345 edn. Menlo Park, California, USA: Google.com; 2001: Freeware. [https://earth.google.com/]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eNovak T: \u003cstrong\u003eGeo Data Viewer - ViewMap GDV-4: geographic data viewer\u0026nbsp;\u003c/strong\u003eIn\u003cem\u003e.\u003c/em\u003e Edited by [github.com/RandomFractals/geo-data-viewer], v 2.6.0 edn. Chicago, Illinois, USA: [marketplace.visualstudio.com/items?itemName=RandomFractalsInc.geo-data-viewer]; 2020: Apache License, V 2.0. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eSchindelin J, Xhmikos R, Oliff C, Contributors.: \u003cstrong\u003eGit Bash For Bash Emulation and Connect In The Secure Shell Protocol (SSH)\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by github.com, v 2.40.0 edn. Global community: gitforwindows.org; 2014: GNU- GP License v2. [github.com/git-for-windows/git/wiki/FAQ/_history?page=3]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eKeyhole Markup Language (KML) : standard of the Open Geospatial Consortium (OGC)\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Keyhole_Markup_Language]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eScalable Vector Graphics (SVG) and Geographic Information System.\u0026nbsp;\u003c/strong\u003e[https://www.directionsmag.com/article/3916]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003eW3C.org: \u003cstrong\u003eScalable Vector Graphics (SVG) Version 2 features and syntax specification\u003c/strong\u003e. In: \u003cem\u003eStandard.\u003c/em\u003e 2018. [https://github.com/w3c/svgwg]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003eButler H, Daly M, Doyle A, Gillies S, Hagen S, Schaub T: \u003cstrong\u003eRFC 7946 - The GeoJSON Format, august 2016 - proposed standard\u003c/strong\u003e. In [rfc-editor.org]; 2016. https://doi.org/10.17487/RFC7946.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eThe GeoJSON Specification (RFC 7946) - a format for encoding a variety of geographic data structures.\u0026nbsp;\u003c/strong\u003e[https://geojson.org/]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGeographic JSON (geojson)\u003c/strong\u003e. In: \u003cem\u003eStandard.\u003c/em\u003e IETF; 2022. [Datatracker.ietf.org]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eShapefile, proprietary format now opened by ESRI as a \u0026apos;standard\u0026apos; with extensions .shp, .shx, .dbf.\u0026nbsp;\u003c/strong\u003e[https://fr.wikipedia.org/wiki/Shapefile]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eShapefile or Geopackage: which format to choose? / \u003cem\u003eShapefile ou Geopackage: quel format choisir?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://geoma-sig.com/2020/09/29/shapefile-ou-geopackage-quel-format-choisir/]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGeospatial Data and GeoJSON and Draw and Edit with GeoJson.io.\u0026nbsp;\u003c/strong\u003e[https://handsondataviz.org/geojson.html; https://handsondataviz.org/geojsonio.html]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003e17.17. Getting Started with the QGIS\u0026apos; Graphical Modeler.\u0026nbsp;\u003c/strong\u003e[https://docs.qgis.org/3.28/fr/docs/training_manual/processing/modeler_twi.html]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eHow do I convert Photoshop ATN files to XML? / \u003cem\u003eComment faire pour convertir des fichiers Photoshop ATN \u0026agrave; XML?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://community.adobe.com/t5/photoshop-ecosystem-discussions/convert-xml-to-atn-action-file/m-p/11935514]. Accessed 16 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eHow do I monitor my computer? / \u003cem\u003eComment je fais ma veille?\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://grafikart.fr/blog/veille-grafikart-organique]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003eKimball S, Mattis P, Mena F, Singh M, Neumann S, Natterer M, Neary D, Nordholts M, Jehan.: \u003cstrong\u003eGIMP: The Free \u0026amp; Open-Source Image Editor\u003c/strong\u003e. V 2.10.34 edn. Berkeley, California, USA: Gimp.org; 1995: CC BY-SA 4.0 License. [Gitlab.gnome.org]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eAndler JA, Bintz J, Marquardt C, B\u0026ouml;ck H, Uro\u0026scaron;ević A, Engelen J, Kiirala N, Taxel P, Hetherington C, Conde-Rueda X\u003cem\u003e\u0026nbsp;et al\u003c/em\u003e: \u003cstrong\u003eInkscape\u003c/strong\u003e. V 1.2.2 edn. Global community: Inkscape.org; 2003: GPLv2 License. [github.com/inkscape/inkscape]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSimple Inkscape Scripting For Inkscape.\u0026nbsp;\u003c/strong\u003e[https://github.com/spakin/SimpInkScr]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003eBogle B, Gundersen E: \u003cstrong\u003eMapbox: a collaborative platform\u003c/strong\u003e. V2.15.0 edn. San Francisco, Californie, USA: Mapbox.com; 2010: Proprietary license. [github.com/mapbox]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eTuto@Mate (SHS - Human Social Science) Thimotee Giraud Package R Cartography / \u003cem\u003eTuto@Mate (SHS - Sciences humaines et sociales) Thimotee Giraud Package R Cartographie.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.youtube.com/watch?v=OI3_AOg6pfc\u0026amp;t=129s]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003eGiraud T, Hernang\u0026oacute;mez D, Pecout H, Ysebaert R, Fellows I, Lemon J, Zenoni F: \u003cstrong\u003eMapsf: Thematic Cartography. R package\u003c/strong\u003e. In: \u003cem\u003eMagrit, cartography.\u003c/em\u003e Edited by [github.com/riatelab/mapsf/], v 0.7.0 edn. Paris, France: 2021: GPL-3. [riatelab.github.io/mapsf/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eGiraud T, Lambert N: \u003cstrong\u003eR package: Cartography\u003c/strong\u003e. V 3.1.3 edn. Paris, France: cloud.r-project.org; 2017: GPL-3. [github.com/riatelab/cartography]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eWagniez P: \u003cstrong\u003ePhicarto guide: Philcarto_2020_MAJ_DOC / \u003cem\u003eGuide de Phicarto: Philcarto_2020_MAJ_DOC\u003c/em\u003e\u003c/strong\u003e; 2019:[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/Philcarto_2020_MAJ_DOC.zip]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003eWagniez P: \u003cstrong\u003ePhilcarto\u0026apos;s guide: DOCdeGRANITn1 / \u003cem\u003eGuide de Philcarto: DOCdeGRANITn1\u003c/em\u003e\u003c/strong\u003e; 2023.[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/DOCdeGRANITn1.zip]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003eWagniez P: \u003cstrong\u003ePhilcarto_2020_QuickStartGuide in 11 lessons for beginners / \u003cem\u003ePhilcarto_2020_GuideDemarrageRapide en 11 le\u0026ccedil;ons pour les d\u0026eacute;butants\u003c/em\u003e\u003c/strong\u003e; 2020:[http://philcarto.free.fr/02_telecharger_logiciels/02_1_telecharger_philcarto/fichiers_installer_philcarto/PhilcartoGuideDemarrage.zip]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eJenks Natural Breaks \u0026mdash; The Best Range Finder algorithm.\u0026nbsp;\u003c/strong\u003e[https://medium.com/analytics-vidhya/jenks-natural-breaks-best-range-finder-algorithm-8d1907192051]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eJenks natural breaks optimization.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/Jenks_natural_breaks_optimization]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eEdit and Join with Mapshaper.\u0026nbsp;\u003c/strong\u003e[https://handsondataviz.org/mapshaper.html#]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eProtection of map data - uMap / \u003cem\u003eProtection des donn\u0026eacute;es cartographiques uMap.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://forum.openstreetmap.fr/t/protection-des-donnees-cartographiques-umap/7251]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eTutorials: creating a personalized map with uMap / \u003cem\u003eTutoriels: cr\u0026eacute;er une carte personnalis\u0026eacute;e avec uMap.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://wiki.cartocite.fr/doku.php?id=umap:tutoriel_umap]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003euMap\u0026quot; user guide / \u003cem\u003eGuide d\u0026apos;utilisation \u0026quot;uMap\u0026quot;\u003c/em\u003e.\u0026nbsp;\u003c/strong\u003e[https://framalibre.org/content/umap]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eImagined Forrest Gump run (1994 film). \u003cem\u003euMap: exemple de diaporama\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[http://umap.openstreetmap.fr/en/map/imagined-forrest-gump-run-1994-film_117755#5/39.113/-93.384]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eMyOSMatic - Online web printing service / \u003cem\u003eService web d\u0026apos;impression en ligne.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://print.get-map.org/]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eAdditional census documentation / \u003cem\u003eDocumentation compl\u0026eacute;mentaire sur le recensement.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://www.insee.fr/fr/information/2383410]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eUnderstanding Legal Populations / \u003cem\u003eComprendre les populations l\u0026eacute;gales.\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e[https://www.insee.fr/fr/information/2553979]. Accessed 27 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eThematic Maps With Mapsf (R Package).\u0026nbsp;\u003c/strong\u003e[https://riatelab.github.io/mapsf/]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003euMap developer documentation -installation on ubuntu - linux.\u0026nbsp;\u003c/strong\u003e[https://umap-project.readthedocs.io/en/master/]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDjango project documentation: Web framework\u0026nbsp;\u003c/strong\u003e[https://docs.djangoproject.com/en/4.2/]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003eBoniface Y: \u003cstrong\u003eumap-project: Add An API For Automatic Export\u003c/strong\u003e. In: \u003cem\u003e/umap/issues/78.\u003c/em\u003e v 1.4.0. edn. France; 2014: WTFPL license. [github.com/umap-project/]. Accessed 28 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eL\u0026apos;API MapOsmatic: uMap GeoJson Export API Support / \u003cem\u003ePrise en charge de l\u0026apos;API MapOsmatic d\u0026apos;exportation Umap GeoJson.\u003c/em\u003e\u0026nbsp;\u003c/strong\u003e[https://blog.get-map.org/index.php/2019/06/16/umap-geojson-export-api-support/]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eForum uMap.\u0026nbsp;\u003c/strong\u003e[https://forum.openstreetmap.fr/c/utiliser/umap/29]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCommunity gitter forum for the umap-project / umap.\u0026nbsp;\u003c/strong\u003e[https://app.gitter.im/#/room/#umap-project_umap:gitter.im]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eMap search on the keyword \u0026quot;health\u0026quot; hosted on the OSM server and shared with the community.\u0026nbsp;\u003c/strong\u003e[https://umap.openstreetmap.fr/fr/search/?q=health]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGPX file format (v 1.1): Open Geospatial Consortium (OGC)\u0026apos; standard.\u0026nbsp;\u003c/strong\u003e[https://en.wikipedia.org/wiki/GPS_Exchange_Format]. Accessed 26 June 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eImporting remote data from a CSV file - Using / \u003cem\u003eImporter des donn\u0026eacute;es distantes depuis un fichier CSV\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e[https://forum.openstreetmap.fr/t/importer-des-donnees-distantes-depuis-un-fichier-csv/2601/3]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eI built a geospatial ETL pipeline with python and this is what I learned.\u0026nbsp;\u003c/strong\u003e[https://medium.com/analytics-vidhya/i-built-a-geospatial-etl-from-scratch-with-python-and-this-is-what-i-learned-b45b37d15f94]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003ePhotoshoplus offers settings, tutorials and colors for Photoshop. And it\u0026apos;s all free!\u0026nbsp;\u003c/strong\u003e[https://www.photoshoplus.fr/]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003eROZIER Guillaume: \u003cstrong\u003eCovidtracker.fr dashboard several analysis scales\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by github.com. Paris, France: OVHCloud; 2020: MIT license. [https://github.com/rozierguillaume/covidtracker-tools]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003eFabacher T: \u003cstrong\u003eCovid-19 outbreak - dynamic atlas, global and national scales / \u003cem\u003eCovid-19 outbreak - atlas dynamique, \u0026eacute;chelles mondiale et nationale\u003c/em\u003e\u003c/strong\u003e. Strasbourg, France; 2020: GMRC-ICUBE-Lab Freeware. [https://teststid.shinyapps.io/covid_19/]. Accessed 17 July 2024.\u003c/li\u003e\n \u003cli\u003eGiraud T, Lambert N: \u003cstrong\u003eReproducible Cartography\u003c/strong\u003e. In: \u003cem\u003eAdvances in Cartography and GIScience: 2017\u003c/em\u003e: Springer Cham; 2017: 173-183. https://doi.org/10.1007/978-3-319-57336-6.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOne year of confinement: some dates that marked the lives of Belgians (chronology) February 26, 2021 / \u003cem\u003eUn an de confinement : quelques dates qui ont marqu\u0026eacute; la vie des Belges (chronologie) 26 f\u0026eacute;vrier 2021\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e[https://www.lespecialiste.be/fr/actualites/un-an-de-confinement-quelques-dates-qui-ont-marque-la-vie-des-belges-chronologie.html?fbclid=IwAR1zxNCGrZteJ29ETcy7Jkk1UKe0oJsQLHrANIbTKBl6IZOYphlk8R6U2zc]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eData engineering in GIS: let the ETL journey begin-ETL - Extract, Transform, Load / \u003cem\u003eIng\u0026eacute;nierie des donn\u0026eacute;es dans les SIG: que le voyage ETL commence-\u003c/em\u003eETL \u0026ndash; Extract, Transform, Load.\u0026nbsp;\u003c/strong\u003e[https://gisgeography.com/data-engineering-gis/]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGithub: Package Cartography.\u0026nbsp;\u003c/strong\u003e[https://github.com/riatelab/cartography]. Accessed 14 July 2024.\u003c/li\u003e\n \u003cli\u003eZirar W, ticsante-com.: \u003cstrong\u003eHospital health data: the Healthcare Data Institute calls for the creation of a \u0026quot;national observatory\u0026quot; / \u003cem\u003eDonn\u0026eacute;es de sant\u0026eacute; hospitali\u0026egrave;res: le Healthcare Data Institute appelle \u0026agrave; la cr\u0026eacute;ation d\u0026apos;un \u0026quot;observatoire national\u0026quot;\u003c/em\u003e\u003c/strong\u003e. \u003cem\u003eTicsantecom\u0026nbsp;\u003c/em\u003e2023, \u003cstrong\u003e81\u003c/strong\u003e.[https://www.ticsante.com/story?ID=6696]. Accessed 14 July 2024\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eThe medical photo library: the first solution for managing medical photos.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ePixacare automates filing and secures photos / \u003cem\u003eLa phototh\u0026egrave;que m\u0026eacute;dicale: premi\u0026egrave;re solution de gestion des photos m\u0026eacute;dicales. Pixacare automatise le classement et s\u0026eacute;curise les photos.\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e[https://en.pixacare.com/a-propos]. Accessed 14 July 2024.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Geomatic, Geographic Mapping, Choropleth Mapping, Collaborative Atlas Platform, Mortality, Spatial Analysis, Alert System, Digital Health, Medical Informatics.","lastPublishedDoi":"10.21203/rs.3.rs-4796017/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4796017/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn the wake of the health crisis, for the year 2020 we have created a France-wide geomatic project to produce several atlases of mortality for all pathologies, with very fine grids at commune and arrondissement levels (Marseille, Lyon, and Paris).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe aim \u003c/strong\u003eis to bring to the collaborative map-sharing platform uMap environment, original visualization and knowledge, and decision-making aids, complementary to existing information and relevant to both the general public and healthcare professionals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethod: \u003c/strong\u003eWe followed a two-step geomatic action plan (with and without webmapping) to create, from each of our four variables, an atlas of the 1\u003csup\u003est\u003c/sup\u003e French containment period. Interpretation of the atlas is facilitated by the graphical display of colored choropleth maps with legend and statistical tooltip, and by rapid transmission to the user of an \"Excess Mortality alert\". In two different webmapping environments tested in advance, our\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eresults \u003c/strong\u003eonly display uMap and its TrafficLight atlas, that instantly transmits the Excess Mortality alert by means of semantic interplay of colors. Comparison between regions is facilitated by the display of administrative contours and by a layer and legend management tool. Today, uMap delivers the expected results without having to worry about its remote server managed by OpenStreetMap, and its interoperability allows atlases to be exported and printed.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOutlook\u003c/strong\u003e: Having taken a step back from geomatic technologies, uMap shows great potential for its implementation dedicated to healthcare on a local server. Further development, to create and associate legends with choropleth maps, could increase uMap’s functionality and interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eWe have innovated, demonstrated and reinforced the value of the collaborative and interoperable uMap mapping platform for visual rendering by instantiating our thematic atlases with choropleth maps and their legends. The TrafficLight atlas is highly relevant for the instant message it communicates by commune, and in this way, we have provided additional information easy to interpretable for all audiences. Finally, uMap offers many advantages in terms of licensing, design, and use of the atlases. This encourages us to continue improving it with the help of its contributors, and to explore it further, with an optimized geomatic action plan.\u003c/p\u003e","manuscriptTitle":"Relevance Of The uMap Collaborative Platform As A Support For Choropleth Mapping: An Atlas Applied To All-Cause Excess Mortality Alerts By Traffic Light - 1st French Containment In 2020","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-09 10:10:45","doi":"10.21203/rs.3.rs-4796017/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e0fd4b1e-acf6-45d2-be9b-b8393d01a44e","owner":[],"postedDate":"August 9th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-08-09T10:10:46+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-09 10:10:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4796017","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4796017","identity":"rs-4796017","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}