Evaluation of state-of-the-art models for improving the diagnosis of mango tree diseases through supervised learning of symptom images in Burkina Faso

Evaluation of state-of-the-art models for improving the diagnosis of mango tree diseases through supervised learning of symptom images in Burkina Faso | 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 Article Evaluation of state-of-the-art models for improving the diagnosis of mango tree diseases through supervised learning of symptom images in Burkina Faso Zoéyandé Oumarou DIANDA, Emile KABORE, Zakaria Cheick Oumar KEITA, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7829136/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract In Burkina Faso, the mango industry faces numerous phytosanitary constraints. This situation is exacerbated by producers' lack of knowledge about diseases and their symptoms, and the low uptake of control technologies. This study aims to develop a tool using artificial intelligence-based approaches for better integrated management of the main diseases affecting mango trees. Surveys conducted in 40 mango orchards yielded 11,001 images of the characteristic symptoms of three major diseases, including 4,527 for anthracnose, 3,038 for bacterial disease and 3,436 for dieback. The data underwent several preprocessing steps. To ensure a balance between disease classes, 3,000 images per class were selected, for a total of 9,000 images. The images were then annotated and used for training with the pre-trained YOLOv11 model. Following training, three models were formed : V2-3M (MAP@50 = 50.3% ; precision = 55.1% ; recall = 49.1%), V3-3M (MAP@50 = 23% ; precision = 56.8% ; recall = 26.7%), and V4-3M (MAP@50 = 46.8% ; precision = 52% ; recall = 45.3%). The most effective model is the V2-3M, as it combines 55.1% accuracy, 50.3% MAP@50 and 49.1% recall. In order to improve diagnosis and phytosanitary management in orchards, it would be useful to integrate treatment recommendations into the model, then deploy it on a mobile application and promote it to producers. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Health sciences/Health care mango tree diseases artificial intelligence automatic diagnosis Burkina Faso Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 INTRODUCTION Mangoes are the fifth most produced fruit in the world (by tonnage) behind bananas/plantains, citrus fruits, grapes and apples, with a production of 51 million tonnes in 2017 1 . Burkina Faso contributes between 11% and 18% of mango production in West Africa 2 . Mangoes are a very important socio-economic and climate issue in Burkina Faso. Mango production and export contribute significantly to the country's income. Indeed, the mango sector generates more than 15 billion in turnover per year 1 . In addition, it contributes to food security in producer areas and vegetation cover, limiting the effects of climate change 2 . The mango value chain in Burkina Faso is a source of employment. Its activities employ 28,000 people in 21,000 jobs as entrepreneurs or similar (farmers, trackers, retailers, etc.) 2 , around 350 permanent jobs in processing and packaging companies for fresh exports, and around 7,000 seasonal jobs throughout the chain 3 . The mango sector faces enormous phytosanitary constraints, particularly insect pests 4 and also major diseases such as anthracnose, bacterial disease and dieback 567 . These diseases cause significant damage in orchards, such as deterioration in fruit quality, tree death and reduced yields 89 . This damage causes enormous losses, such as loss of income for producers, management costs, reduced productivity, etc. 10 . Unfortunately, producers have little understanding of these diseases and their symptoms, epidemiological aspects and diagnostic difficulties. Added to this is the limited dissemination of available control technologies, resulting in low adoption by producers. Nowadays, artificial intelligence is a powerful decision-making tool in agriculture, particularly in the automatic diagnosis of crop diseases 11 . In India, automatic diagnosis and classification models for crop pests have been implemented 1213 . This aspect of AI for agricultural development is still largely unexplored in Burkina Faso, specifically for the diagnosis of mango tree diseases. The aim of this study is to design an effective AI-based automatic diagnosis model for the main mango tree diseases in Burkina Faso. MATERIAL AND METHODS Study area The study was conducted in the Guiriko (Houet and Kénédougou provinces) and Tannounyan (Comoé and Léraba provinces) regions (Figure 1), which account for approximately 70 % of national mango production. Orchards in this area are affected by major diseases such as dieback 14 , bacterial disease 6 and anthracnose 5 . The local economy is based on agriculture, livestock farming and trade, with the mango industry supported by industrial and artisanal processing units. 15 16 Collection of image data on diseases in mango orchards Surveys, collection and compilation of representative images Surveys were conducted in 10 orchards in each of the four provinces studied (Kénédougou, Houet, Comoé and Léraba), for a total of 40 mango orchards. The selection was based on the accessibility of the orchard and the consent of the producer. Careful observations were made in the orchards. These enabled the identification of mango trees showing symptoms of the three main diseases, namely dieback, anthracnose and bacterial disease. In the selected orchards, photographs of the symptoms of the main diseases were taken with smartphones. The image collection operation mainly focused on the leaves, fruit and the entire tree in cases of severe decline. The images were taken during the day and from several angles of the organs affected by the diseases in order to obtain a better representation of the different symptoms. The images collected were stored on an external hard drive by disease class. Pre-processing, image annotation The images were pre-processed to improve their quality and usability. This mainly involved resizing them to dimensions compatible with YOLOv11, normalising colours and correcting brightness, and increasing the data (data augmentation) through rotation, inversion, cropping and noise to improve the robustness of the model. After pre-processing, batches of 3,000 images for each disease were selected, for a total of 9,000 images for all three diseases (anthracnose, bacterial disease and dieback). These selected images were imported into the Roboflow platform for annotation. Image annotation consisted of drawing bounding boxes around the symptoms and associating them with a class corresponding to each disease (Figure 2). This methodical work made it possible to build a systematically annotated dataset that will be used for training. Image training Supervised transfer learning with the pre-trained YOLOv11 algorithm was used. YOLOv11 (You Only Look Once, version 11) is a convolutional neural network (CNN) designed for real-time object detection and simultaneously enables : • Automatic extraction of relevant visual features, • Localisation of symptoms via bounding boxes, • Classification of detected diseases. The entire supervised training process was conducted via the Roboflow platform, which provides an integrated environment for managing datasets and training computer vision models. The annotated data was divided into three sets : 75 % for training (training set), 20 % for validation (validation set), and 5 % for testing (test set). Regularisation techniques such as early stopping were applied to avoid overfitting. All data was made available on a platform for easy access. Model performance was evaluated using standard object detection metrics such as: Precision : the proportion of correct predictions among all predictions made. It indicates the model's ability to limit false detections. Recall : this is the proportion of cases correctly detected in relation to all existing positive cases. It reflects the model's ability to not omit any symptoms that are present. mAP@50 (mean Average Precision at 50% IoU): this is an overall indicator of object detection performance, calculated as the average precision across all classes, with an intersection over union (IoU) threshold set at 50%. These metrics were used to evaluate the models' ability to correctly identify and locate symptoms of mango tree diseases, while ensuring a balance between accuracy and completeness. Data processing and analysis The data was organised and classified in Excel 2019. Excel was used to total the number of images per disease and to create graphs. The GIS tool (My Coordinates) was used to record the geographical coordinates of the various sites for mapping purposes using QGIS software. Robboflow software was used for image annotation and model training. RESULTS Symptoms of the three main annotated diseases Surveys carried out in the four provinces enabled the symptoms of anthracnose, bacterial disease and dieback to be identified and annotated using the Roboflow platform. Anthracnose is characterised by black spots that develop into necrosis on the leaves and deep necrotic lesions on the fruit (Figure 3). Bacterial disease manifests as irregular oily spots on the leaves, sometimes perforated, and blackish star-shaped and cracked lesions on the fruit, sometimes with bacterial exudate (Figure 4). The dieback caused the leaves to wilt and gradually turn brown, with partial (10 to 90%) and total (100%) coverage (Figure 5). Performance of automatic detection models for major diseases. Training identified three model variants named V2-3M, V3-3M and V4-3M. These models were evaluated based on the performance of standard metrics, namely precision, recall and Mean Average Precision at an IoU of 50 % (mAP@50) (Figure 6). The evaluation reveals contrasting performances in the detection of symptoms of anthracnose, bacterial disease and mango tree decline. The V2-3M model stands out with its best results, with a mAP@50 of 50,3 %, precision of 55,1 % and recall of 49,1 %. The V4-3M model performs slightly worse, with an mAP@50 of 46.8%, an accuracy of 52 % and a recall of 45,3 %. In contrast, the V3-3M model shows the lowest results, with an mAP@50 of 23,5 % and a recall of only 26,7 %, despite relatively good precision (56,8 %). DISCUSSION Distinctive symptoms of mango tree diseases: a solid basis for establishing an automatic diagnostic system. Accurate identification of phytopathological symptoms is a fundamental prerequisite for the development of digital diagnostic tools, particularly those using AI. The symptoms documented in orchards largely corroborate the classic descriptions established by researchers. Indeed, for anthracnose, observations of irregular brown spots on leaves, sunken black lesions on fruit and necrosis of young twigs are consistent with the work of 17 and 18 , which highlight similar symptoms of anthracnose in humid mango-producing areas, where relative humidity and poor aeration promote infection. The work of 19 has also demonstrated the early appearance of symptoms as early as the flowering stage, directly influencing yield. With regard to bacterial disease, the photographed symptoms captured, such as translucent oily spots, leaf perforations and exudates on fruit, are typical of infection by Xanthomonas campestris pv. mangiferaeindicae. These symptoms were described by 6 in Burkina Faso and West Africa, who highlighted the rapid progression of bacterial lesions. Recognition of these visual characteristics is crucial in order to differentiate this disease from anthracnose, whose lesions can sometimes overlap on the fruit. With regard to decline, symptomatic analysis reveals leaf wilting, progressive browning of twigs, and fruit abortion 20 . This description corresponds to the symptoms associated with complex fungal infections, particularly Lasiodiplodia spp , as reported by 21 . All of this symptomatic data thus constitutes a consistent database for training AI models, particularly through deep learning, to identify major diseases. This has been demonstrated in the work of 22 on the automatic identification of wheat diseases in Europe using AI algorithms. Also, based on the training of approximately 8,438 images of symptoms acquired on PlantVillage, 23 have set up a system for the automatic detection of diseases in vineyards and mango trees in India. This will constitute a database of the symptoms of the main diseases affecting mango trees and, above all, a significant step forward in overcoming the lack of phytosanitary expertise and speeding up decision-making in the field by producers. Best model for automatic detection of major mango tree diseases. The performance achieved by the V2-3M, V3-3M and V4-3M models demonstrates differentiated capabilities in detecting the symptoms of major mango tree diseases from annotated images with accuracies greater than 50 %. The V2-3M model, with an mAP@50 of 50.3%, appears to be the most effective, confirming the importance of good parameterisation and optimised architecture in automatic detection tasks. Its performance combines relatively balanced precision (55,1 %) and recall (49,1 %), indicating that it can be used as a starting point for a functional prototype of a mobile plant health diagnosis tool 24 . These metrics show a good compromise between the model's ability to correctly locate symptoms and its ability to minimise both false positives and false negatives. On the other hand, the poor performance of the V3-3M model (mAP@50 = 23,5 % ; recall = 26,7 %) despite similar accuracy (56,8 %) highlights limitations in the generalisation of learning. This model seems to struggle to capture symptomatic diversity, particularly due to reduced sensitivity to morphological and chromatic variations in symptoms. The metrics of the V4-3M model, which are intermediate (mAP@50 = 46,8 %) and reflect a certain robustness, albeit less marked than V2-3M, show that improved performance depends on a compromise between architectural complexity, annotation quality and training image diversity. This finding is consistent with the observations of 25 , who showed that training models on highly diversified image databases significantly improves the robustness of artificial intelligence tools applied to agriculture. Furthermore, the satisfactory performance of the V2-3M model is in line with recent research into leaf disease detection using deep learning. For example, 26 achieved a mAP of 51% on tomato diseases using CNN architectures. 12 achieved 99 % accuracy for a cashew anthracnose diagnosis model using tools such as Mobilenet and Inception. In terms of performance, the V2-3M model is the most effective. Its ability to balance accuracy (55,1 %) and recall (49,1 %), combined with an mAP@50 of 50,3 %, makes it the most suitable for initial operational application. The V4-3M model, although fairly close, remains slightly less robust, while V3-3M should be ruled out due to its insufficient results. Thus, V2-3M provides a solid basis for an automated diagnostic system, with prospects for improvement. These results confirm the feasibility of using artificial intelligence models for the automatic detection of major mango tree diseases, while highlighting the need to continue optimising algorithms and enriching training databases 27 . The integration of such a tool into a mobile application could strengthen early detection capabilities and improve the responsiveness of phytosanitary interventions in Burkina Faso. CONCLUSION Anthracnose, bacterial disease and dieback are major phytosanitary constraints affecting mango trees. Unfortunately, producers have little knowledge of the epidemiological aspects of these diseases or the control techniques available. This study was conducted with the aim of designing an accurate automatic diagnostic model for the main diseases affecting mango trees using artificial intelligence. Training the collection of symptomatic images showed that the V2-3M model performed better in terms of metrics for automatic disease detection. These relevant results open up new prospects for integrating AI tools into agroecological practices, particularly for phytosanitary monitoring and management in orchards. It would be wise to integrate treatment recommendations into the model, deploy it on a mobile application, and promote it for the benefit of producers, all of which will contribute to effective disease management in mango orchards. Declarations FUNDING This work benefited from the facilities of ACTS/RIMP/FONRID/INERA-01/2023 fund ACKNOWLEDGEMENTS The research was conducted at the Regional Center of Excellence in Fruits and Vegetables (CRE/FL) based at the Institute of Environment and Agronomic Research (INERA) in Farako-bâ, Burkina Faso. Authors' contributions DZO : Conceptualization; Funding acquisition; Methodology; Resources; Roles/Writing - original draft, KE : Manuscript’s writing , DZO, KE, ZI, KZCO : Data collection, KE and KZCO : Initial processing, data treatement and analysis, ZI, KZ, LB, MS, WI : Critical revision and improvement of the scientific content All authors have read and approved the final version of the manuscript. Data availability statement (mandatory) Data from this study will be made available to the journal as needed via the corresponding author's e-mail address. Conflict of interest The authors certify that no competing interests or conflicts of interest, whether institutional, financial or personal, have interfered with the scientific integrity, methodology used or objectivity of the results presented in this article. References Food and agriculture organization (FAO) - Statistical Yearbook . FAO Rome, (2021). APEX. Fiche Sectorielle: Mangue Du Burkina Faso . (2018). https://www.apexb.bf/catalogue-de-produits/mangue? doi:https://www.apexb.bf/catalogue-de-produits/mangue? Ministry of Agriculture and Halieutic Resources (MARAH). Filière Mangue: Les Acteurs Font Le Bilan de La Campagne.. (2023). Zida, I. et al. Effectiveness of Four Integrated Pest Management Approaches in the Control of Fruit Flies (Diptera: Tephritidae) in Mango Agro-Ecosystems in the South-Sudanian Zone of Burkina Faso. Adv. Ento- mology . 11 , 124–142 (2023). Bougoum, H. et al. Caractérisation de Colletotrichum spp agent pathogène de l ’ anthracnose du manguier (Mangifera indica L.) au Burkina Faso. J. Appl. Biosci. 190 , 19999–20020 (2023). Zombré, C. diversite genetique et pathologique de xanthomonas citripv. mangiferaeindicae, bacterie responsable de la maladie des taches noires du manguier au burkina faso et en afrique de l’ouest » (Université Joseph Ki-Zerbo, Ouagadodugou, 2016). Dianda, Z. O. et al. Prévalence du dessèchement du manguier et evaluation de la fréquence des champignons associés à la maladie au Burkina Faso. J. Appl. Biosci. 126 , 12686 (2018). Dembele, D. D. & Camara, B. Evaluation pré et post-récolte de l ’ incidence et de la sévérité de l ’ anthracnose du manguier au nord de la Côte d ’ Ivoire. ResearchGate 1–2 doi: (2019). https://www.researchgate.net/publication/341314189 Paul-alfred, K. K. Profitability and Competitiveness of the Mango Sector in Ivory Coast. Int. J. Econ. Financ Res. 10 , 12–16 (2024). PIP COLEACP. Nouveaux ravageurs et maladies invasives. Bactériose du manguier provoquée par Xanthomonas citri pv. mangiferaeindicae. COLEACP PIP 1–14 p. doi: (2012). Lahrache, R. L ’ impact de l ’ intelligence artificielle sur la prise de décision The Impact of Artificial Intelligence in Decision Making. 7, 660–678 (2024). Jayaprakash, V. & Rajagopal, M. K. cashew dataset generation using augmentation and ralsgan and a transfer learning based tinyml approach abstract: arXiv Prepr. arXiv2304.08766 (2023). Tirkey, D., Singh, K. K. & Tripathi, S. Performance analysis of AI-based solutions for crop disease identification, detection, and classification. Smart Agric. Technol. 5 , 100238 (2023). Dianda, O. et al. Prévalence du dessèchement du manguier et evaluation de la fréquence des champignons associés à la maladie au Burkina Faso. J. Appl. Biosci. 12686–12699. https://dx.doi.org/10.4314/jab.v126i1.6 (2018). INSD. Profil de Pauvreté de La Région Des Hauts Bassins. (2023). INSD. 5è Recensement Général de La Population et de l’Habitat . (2022). doi:www.insd.bf Dembele, D. D., Og, L., Georges, D. & Amari, E. Pre and postharvest assessment of mango anthracnose incidence and severity in the north of Côte d ’ Ivoire Pre and postharvest assessment of mango anthracnose incidence and severity in the north of Côte d ’ Ivoire. 14 p. (2020). 10.4314/ijbcs.v13i6.24 Dofuor, A. K. et al. Mango anthracnose disease: the current situation and direction for future research. Front. Microbiol. 14 , 1–18 (2023). Kankam, F. & Adomako, J. Anthracnose Disease of Mango: Epidemiology, Impact and Management Options. (2022). 10.5772/intechopen.105934 Dianda, Z. O. et al. P. Prévalence du dessèchement du manguier et evaluation de fréquence des champignons associés à la maladie au Burkina Faso. J. Appl. Biosci. 12686–12699. https://dx.doi.org/10.4314/jab.v126i1.6 (2018). Oumarou Zoéyandé Dianda1, Issa Wonni1, Léonard Ouédraogo1, Philippe Sankara2, Charlotte 4 Tollenaere1, 3, Emerson M. Del Ponte4, D. F. Lasiodiplodia species associated with mango ( Mangifera indica L.) decline in Burkina Faso and influence of climatic factors on the disease distribution. (2023). Picon, A. & Alvarez-Gila, A. C, M. S. & Amaia Ortiz-Barredo, Jone Echazarra, A. J. Réseaux neuronaux convolutifs profonds pour la classification des maladies des cultures à l’état sauvage à partir d’appareils de capture mobiles. ScienceDirect 161, 280–290 (2020). Rao, U. S., Swathi, R., Sanjana, V., Arpitha, L. & Chandrasekhar, K. Deep Learning Precision Farming: Grapes and Mango Leaf Disease Detection by Transfer Learning. Glob. Transitions Proc. 2, 535-544p (2021). Elkadi, K. Expérimentation d’un modèle de détection précoce des maladies de la tomate par apprentissage profond. Rev. Marocaine Prot. des. Plantes . 14 , 19–30 (2020). Ramcharan, A., Mccloskey, P., Baranowski, K. & Mbilinyi, N. A. Mobile-Based Deep Learning Model for Cassava Disease Diagnosis. Front. Plant. Sci. 10 , 1–8 (2019). Boulent, J., Foucher, S. & Théau, J. Convolutional Neural Networks for the Automatic Identification of Plant Diseases. Front. Plant. Sci. 10 , 15 (2019). Hossain, M. A., Sakib, S., Abdullah, H. M. & Arman, S. E. Deep learning for mango leaf disease identification: A vision transformer perspective. Heliyon 10 , 36361 (2024). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 11 May, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviews received at journal 13 Feb, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers agreed at journal 18 Jan, 2026 Reviewers agreed at journal 17 Jan, 2026 Reviewers invited by journal 02 Nov, 2025 Editor invited by journal 13 Oct, 2025 Editor assigned by journal 12 Oct, 2025 Submission checks completed at journal 12 Oct, 2025 First submitted to journal 10 Oct, 2025 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-7829136","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":538713661,"identity":"7f2a0d1a-7a93-4d05-a698-085e58d38cd3","order_by":0,"name":"Zoéyandé Oumarou DIANDA","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYFCDw0DMw2DDA+YwNsgRrSUNpsWYCC0HwFoOMxDUwi+Rnfi5ouIwA99x5mcP3tSclzFn7334gXGHAU4tkjNyN0ueOXOYQfIwm7nhnGO3eSx7jhtLMJ7BrcXgRu4Gyca2wwwGhxnMpHnYbvMY3EhjkGBs+4NPy+afEC3s36R5/p3jMbj/jPkHYxteW7ZBbeExk+ZtOwC0hY1NAp8WyZ632ywbzqTzSB7mKZOc25fMY3Amjc0iEY9f+NlzN99sqLCW4zt/fJvEm2929gbHjzHf+IgnxGCAB5WbQFDDKBgFo2AUjAJ8AAAXl1J116zQxwAAAABJRU5ErkJggg==","orcid":"","institution":"National Center for Scientific and Technological Research","correspondingAuthor":true,"prefix":"","firstName":"Zoéyandé","middleName":"Oumarou","lastName":"DIANDA","suffix":""},{"id":538713662,"identity":"41b84166-de32-48a7-b7d7-7f87b05a0a71","order_by":1,"name":"Emile KABORE","email":"","orcid":"","institution":"Yembila Abdoulaye Toguyeni University (UYAT)","correspondingAuthor":false,"prefix":"","firstName":"Emile","middleName":"","lastName":"KABORE","suffix":""},{"id":538713663,"identity":"e5970dd0-1f5f-462f-89e3-0fb7dffc7738","order_by":2,"name":"Zakaria Cheick Oumar KEITA","email":"","orcid":"","institution":"Nazi Boni University (UNB)","correspondingAuthor":false,"prefix":"","firstName":"Zakaria","middleName":"Cheick Oumar","lastName":"KEITA","suffix":""},{"id":538713665,"identity":"07ab6d64-1271-4820-a142-9d073cea404c","order_by":3,"name":"Issaka ZIDA","email":"","orcid":"","institution":"Nazi Boni University (UNB)","correspondingAuthor":false,"prefix":"","firstName":"Issaka","middleName":"","lastName":"ZIDA","suffix":""},{"id":538713668,"identity":"45d46459-7ca1-49af-997b-46d9fc73b001","order_by":4,"name":"Zakaria KINDA","email":"","orcid":"","institution":"Nazi Boni University (UNB)","correspondingAuthor":false,"prefix":"","firstName":"Zakaria","middleName":"","lastName":"KINDA","suffix":""},{"id":538713669,"identity":"927f3aee-7faa-4a53-a624-689507ac0db4","order_by":5,"name":"Lankoandé Benjamin","email":"","orcid":"","institution":"Yembila Abdoulaye Toguyeni University (UYAT)","correspondingAuthor":false,"prefix":"","firstName":"Lankoandé","middleName":"","lastName":"Benjamin","suffix":""},{"id":538713670,"identity":"759f208b-a112-4c11-b60a-4c3f4e8914ed","order_by":6,"name":"Malo Sadouanouan","email":"","orcid":"","institution":"Nazi Boni University (UNB)","correspondingAuthor":false,"prefix":"","firstName":"Malo","middleName":"","lastName":"Sadouanouan","suffix":""},{"id":538713671,"identity":"586c6587-0f0f-45d8-a922-1a5c449fa358","order_by":7,"name":"Issa WONNI","email":"","orcid":"","institution":"National Center for Scientific and Technological Research","correspondingAuthor":false,"prefix":"","firstName":"Issa","middleName":"","lastName":"WONNI","suffix":""}],"badges":[],"createdAt":"2025-10-10 16:53:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7829136/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7829136/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95807125,"identity":"ebeaca95-1320-40ee-aeba-e386fee17664","added_by":"auto","created_at":"2025-11-13 08:48:07","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13299366,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/718349606af461d9d44110f5.tif"},{"id":95806663,"identity":"a7dbf41f-44cc-47f9-9aa0-39851c2b46ce","added_by":"auto","created_at":"2025-11-13 08:47:46","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51936,"visible":true,"origin":"","legend":"","description":"","filename":"AdvancingMangodiseaseDiagnosisDIANDAZ.Oumarou..docx","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/dd043107a4aa48509fa43b46.docx"},{"id":95806702,"identity":"fdd64b62-1f09-45d1-bd63-222cf07de5f9","added_by":"auto","created_at":"2025-11-13 08:47:48","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19133956,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/68ae88acc4804e0fd1eaa943.tif"},{"id":95807127,"identity":"a96b2161-7d72-44c2-b900-89ed31cc457b","added_by":"auto","created_at":"2025-11-13 08:48:07","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34510776,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/68b3fb965deb2cb5eb470eff.tif"},{"id":95807195,"identity":"a2c9579f-00bf-46be-ad40-f5838c4b2616","added_by":"auto","created_at":"2025-11-13 08:48:11","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27139820,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/3f5856782e03782b90460f24.tif"},{"id":95807026,"identity":"3ba6ab41-3ed5-426d-96a0-3a0bbbbbc094","added_by":"auto","created_at":"2025-11-13 08:48:03","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43679628,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/d794c7e7b2fdd5723e288faa.tif"},{"id":95806760,"identity":"cc5c9faa-b360-46e4-a26c-9e3223bd02e1","added_by":"auto","created_at":"2025-11-13 08:47:52","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19199124,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/8c3d3e7e5dea29d9f3cbbfd4.tif"},{"id":95807132,"identity":"e0d599cf-b7f0-401e-8f19-989a6ca91123","added_by":"auto","created_at":"2025-11-13 08:48:08","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14538,"visible":true,"origin":"","legend":"","description":"","filename":"Figures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/294d6ef12cd746151a32f606.docx"},{"id":95807165,"identity":"2758cd2e-047e-4a40-a397-7bd467407961","added_by":"auto","created_at":"2025-11-13 08:48:10","extension":"json","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8935,"visible":true,"origin":"","legend":"","description":"","filename":"fe5f813e1f4740929ea7bbd714ef1cde.json","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/2b195e4afce82de86b05b458.json"},{"id":95806696,"identity":"c6a281ea-1725-4ffa-aa8f-4c71853114eb","added_by":"auto","created_at":"2025-11-13 08:47:48","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":61667,"visible":true,"origin":"","legend":"","description":"","filename":"fe5f813e1f4740929ea7bbd714ef1cde1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/196ddb091735fb8922ac6442.xml"},{"id":95807251,"identity":"20e9800e-f7ab-4aad-adaa-8fd6c5250e43","added_by":"auto","created_at":"2025-11-13 08:48:15","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13299366,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/9d213fbdbf0da928d4965edf.tif"},{"id":95807094,"identity":"99373557-4a52-4bd3-b78c-2981862c006e","added_by":"auto","created_at":"2025-11-13 08:48:06","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19133956,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/7b2841c6327a6d7fe901d7f0.tif"},{"id":95806777,"identity":"49db774a-0461-47ef-af88-3767d861d0e4","added_by":"auto","created_at":"2025-11-13 08:47:52","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34510776,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/6c0385cb429c974d880d976f.tif"},{"id":95806627,"identity":"6fd4ce68-edb0-49b8-ab8f-2bf1a92d3f23","added_by":"auto","created_at":"2025-11-13 08:47:45","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27139820,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/0aa36db246e2ce28258fa8cb.tif"},{"id":95806999,"identity":"3b8aacaa-1a27-406a-a588-8acca5402319","added_by":"auto","created_at":"2025-11-13 08:48:02","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43679628,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/9554b1ada5a9e861bf22ec03.tif"},{"id":95806639,"identity":"89faff8a-569d-454d-b3be-8cb44f5c2280","added_by":"auto","created_at":"2025-11-13 08:47:46","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19199124,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.tif","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/12c9212465c0bd96980b9d22.tif"},{"id":95806761,"identity":"621c6f64-d055-4dc2-9499-e15be96031c6","added_by":"auto","created_at":"2025-11-13 08:47:52","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":960351,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/3b4098ee851cd67ba7e38919.png"},{"id":95807268,"identity":"f149d9f1-936b-4754-86f3-0ce613f4b6c8","added_by":"auto","created_at":"2025-11-13 08:48:15","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1049777,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/ffd6d6655a87370f1cafe7b3.png"},{"id":95807276,"identity":"72a18364-562d-4876-a53b-4c358aabe565","added_by":"auto","created_at":"2025-11-13 08:48:16","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2500337,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/02fb55c8dd32efbe436af045.png"},{"id":95807037,"identity":"9099582d-9382-4220-9658-2bd5d3010474","added_by":"auto","created_at":"2025-11-13 08:48:04","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1883683,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/3ab7f8c00944b58c882356aa.png"},{"id":95807009,"identity":"3fb84f58-a020-4f76-98f2-59d95f40acca","added_by":"auto","created_at":"2025-11-13 08:48:03","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3951905,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/bbd8b5b85139182ca231fd8c.png"},{"id":95807168,"identity":"aadebada-d8bf-4dab-874c-6b17685cf50e","added_by":"auto","created_at":"2025-11-13 08:48:10","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1178873,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/d52783981e5f1e22101b09cd.png"},{"id":95807292,"identity":"52f9e19f-0898-48b5-9066-f23d305c4f1c","added_by":"auto","created_at":"2025-11-13 08:48:16","extension":"xml","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60991,"visible":true,"origin":"","legend":"","description":"","filename":"fe5f813e1f4740929ea7bbd714ef1cde1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/778838d7a8f9483e9d0f2d16.xml"},{"id":95806782,"identity":"476c9414-309b-4b5d-9b12-4448b0ab30f0","added_by":"auto","created_at":"2025-11-13 08:47:53","extension":"html","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":70776,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/63f9d4aa95062a5815104f71.html"},{"id":95807074,"identity":"376dc727-0a64-4c59-8497-83274cc108b6","added_by":"auto","created_at":"2025-11-13 08:48:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":7875909,"visible":true,"origin":"","legend":"\u003cp\u003eMap showing the location of sites in the study area\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/4311d2e1b8d1bb0948ccee24.png"},{"id":95807128,"identity":"86df24ac-7e94-4108-b965-1a5b5fada92a","added_by":"auto","created_at":"2025-11-13 08:48:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":10772936,"visible":true,"origin":"","legend":"\u003cp\u003eAnnotation operation on the Robboflow platform\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/584f8f222c09ca063a1ae3c8.png"},{"id":95806908,"identity":"98189949-e3f8-4cf7-817f-fe2cece432c6","added_by":"auto","created_at":"2025-11-13 08:47:59","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":15931009,"visible":true,"origin":"","legend":"\u003cp\u003eAnnotated symptoms of anthracnose on mango organs.\u003c/p\u003e\n\u003cp\u003eLegend : A and (B): on leaves; (C): on fruit\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/4759ea86c260db849223c33b.png"},{"id":95806738,"identity":"5c9d4786-3835-41d1-ae3f-54ad7458d7cc","added_by":"auto","created_at":"2025-11-13 08:47:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":12889849,"visible":true,"origin":"","legend":"\u003cp\u003eAnnotated symptoms of bacterial blight on mango organs\u003c/p\u003e\n\u003cp\u003eLegend : (A): on leaves (B) and (C): on fruit\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/0dcd960221589793ba25745c.png"},{"id":95807043,"identity":"aaf3233c-a5ae-4af4-b26c-af6d9bb77e00","added_by":"auto","created_at":"2025-11-13 08:48:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":27107288,"visible":true,"origin":"","legend":"\u003cp\u003eAnnotated symptoms of dieback on mango organs\u003c/p\u003e\n\u003cp\u003eLegend : \u0026nbsp;(A): on leaves, (B): partial dieback (D): total dieback.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/39251086a338ee0a60507384.png"},{"id":95806903,"identity":"260dfd51-10c6-4a73-a541-6d33fd371b09","added_by":"auto","created_at":"2025-11-13 08:47:59","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":22471235,"visible":true,"origin":"","legend":"\u003cp\u003eModel characteristics after symptom training\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/194afe848bf40cf3420718a3.png"},{"id":95803999,"identity":"56121c06-7dc0-4525-9c04-04c3120b7cb7","added_by":"auto","created_at":"2025-11-13 08:35:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":620578,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7829136/v1/0896a698-cc33-473e-bffd-6e9ada6b8798.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Evaluation of state-of-the-art models for improving the diagnosis of mango tree diseases through supervised learning of symptom images in Burkina Faso","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eMangoes are the fifth most produced fruit in the world (by tonnage) behind bananas/plantains, citrus fruits, grapes and apples, with a production of 51\u0026nbsp;million tonnes in 2017 \u003csup\u003e1\u003c/sup\u003e. Burkina Faso contributes between 11% and 18% of mango production in West Africa \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Mangoes are a very important socio-economic and climate issue in Burkina Faso. Mango production and export contribute significantly to the country's income.\u003c/p\u003e\u003cp\u003eIndeed, the mango sector generates more than 15\u0026nbsp;billion in turnover per year\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. In addition, it contributes to food security in producer areas and vegetation cover, limiting the effects of climate change \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. The mango value chain in Burkina Faso is a source of employment. Its activities employ 28,000 people in 21,000 jobs as entrepreneurs or similar (farmers, trackers, retailers, etc.) \u003csup\u003e2\u003c/sup\u003e, around 350 permanent jobs in processing and packaging companies for fresh exports, and around 7,000 seasonal jobs throughout the chain \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe mango sector faces enormous phytosanitary constraints, particularly insect pests \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e and also major diseases such as anthracnose, bacterial disease and dieback\u003csup\u003e567\u003c/sup\u003e. These diseases cause significant damage in orchards, such as deterioration in fruit quality, tree death and reduced yields\u003csup\u003e89\u003c/sup\u003e. This damage causes enormous losses, such as loss of income for producers, management costs, reduced productivity, etc.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Unfortunately, producers have little understanding of these diseases and their symptoms, epidemiological aspects and diagnostic difficulties. Added to this is the limited dissemination of available control technologies, resulting in low adoption by producers. Nowadays, artificial intelligence is a powerful decision-making tool in agriculture, particularly in the automatic diagnosis of crop diseases \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. In India, automatic diagnosis and classification models for crop pests have been implemented \u003csup\u003e1213\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThis aspect of AI for agricultural development is still largely unexplored in Burkina Faso, specifically for the diagnosis of mango tree diseases. The aim of this study is to design an effective AI-based automatic diagnosis model for the main mango tree diseases in Burkina Faso.\u003c/p\u003e"},{"header":"MATERIAL AND METHODS","content":"\u003cp\u003e\u003cstrong\u003eStudy area\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted in the Guiriko (Houet and Kénédougou provinces) and Tannounyan (Comoé and Léraba provinces) regions (Figure 1), which account for approximately 70 % of national mango production. Orchards in this area are affected by major diseases such as dieback\u0026nbsp;\u003csup\u003e14\u003c/sup\u003e, bacterial disease\u0026nbsp;\u003csup\u003e6\u003c/sup\u003e and anthracnose\u003csup\u003e5\u003c/sup\u003e. The local economy is based on agriculture, livestock farming and trade, with the mango industry supported by industrial and artisanal processing units.\u003csup\u003e15\u003c/sup\u003e\u003csup\u003e16\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCollection of image data on diseases in mango orchards\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurveys, collection and compilation of representative images\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSurveys were conducted in 10 orchards in each of the four provinces studied (Kénédougou, Houet, Comoé and Léraba), for a total of 40 mango orchards. The selection was based on the accessibility of the orchard and the consent of the producer. Careful observations were made in the orchards. These enabled the identification of mango trees showing symptoms of the three main diseases, namely dieback, anthracnose and bacterial disease. In the selected orchards, photographs of the symptoms of the main diseases were taken with smartphones.\u003c/p\u003e\n\u003cp\u003eThe image collection operation mainly focused on the leaves, fruit and the entire tree in cases of severe decline. The images were taken during the day and from several angles of the organs affected by the diseases in order to obtain a better representation of the different symptoms. The images collected were stored on an external hard drive by disease class.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePre-processing, image annotation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe images were pre-processed to improve their quality and usability. This mainly involved resizing them to dimensions compatible with YOLOv11, normalising colours and correcting brightness, and increasing the data (data augmentation) through rotation, inversion, cropping and noise to improve the robustness of the model. After pre-processing, batches of 3,000 images for each disease were selected, for a total of 9,000 images for all three diseases (anthracnose, bacterial disease and dieback). These selected images were imported into the Roboflow platform for annotation.\u003c/p\u003e\n\u003cp\u003eImage annotation consisted of drawing bounding boxes around the symptoms and associating them with a class corresponding to each disease (Figure 2). This methodical work made it possible to build a systematically annotated dataset that will be used for training.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImage training\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupervised transfer learning with the pre-trained YOLOv11 algorithm was used. YOLOv11 (You Only Look Once, version 11) is a convolutional neural network (CNN) designed for real-time object detection and simultaneously enables :\u003c/p\u003e\n\u003cp\u003e• \u0026nbsp; \u0026nbsp;Automatic extraction of relevant visual features,\u003c/p\u003e\n\u003cp\u003e• \u0026nbsp; \u0026nbsp; Localisation of symptoms via bounding boxes,\u003c/p\u003e\n\u003cp\u003e• \u0026nbsp; \u0026nbsp; Classification of detected diseases.\u003c/p\u003e\n\u003cp\u003eThe entire supervised training process was conducted via the Roboflow platform, which provides an integrated environment for managing datasets and training computer vision models. The annotated data was divided into three sets : 75 % for training (training set), 20 % for validation (validation set), and 5 % for testing (test set). Regularisation techniques such as early stopping were applied to avoid overfitting. All data was made available on a platform for easy access.\u003c/p\u003e\n\u003cp\u003eModel performance was evaluated using standard object detection metrics such as:\u003c/p\u003e\n\u003cp\u003ePrecision : the proportion of correct predictions among all predictions made. It indicates the model's ability to limit false detections.\u003c/p\u003e\n\u003cp\u003eRecall : this is the proportion of cases correctly detected in relation to all existing positive cases. It reflects the model's ability to not omit any symptoms that are present.\u003c/p\u003e\n\u003cp\u003emAP@50 (mean Average Precision at 50% IoU): this is an overall indicator of object detection performance, calculated as the average precision across all classes, with an intersection over union (IoU) threshold set at 50%.\u003c/p\u003e\n\u003cp\u003eThese metrics were used to evaluate the models' ability to correctly identify and locate symptoms of mango tree diseases, while ensuring a balance between accuracy and completeness.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData processing and analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data was organised and classified in Excel 2019. Excel was used to total the number of images per disease and to create graphs. The GIS tool (My Coordinates) was used to record the geographical coordinates of the various sites for mapping purposes using QGIS software. Robboflow software was used for image annotation and model training.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003eSymptoms of the three main annotated diseases\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSurveys carried out in the four provinces enabled the symptoms of anthracnose, bacterial disease and dieback to be identified and annotated using the Roboflow platform.\u003c/p\u003e\n\u003cp\u003eAnthracnose is characterised by black spots that develop into necrosis on the leaves and deep necrotic lesions on the fruit (Figure 3).\u003c/p\u003e\n\u003cp\u003eBacterial disease manifests as irregular oily spots on the leaves, sometimes perforated, and blackish star-shaped and cracked lesions on the fruit, sometimes with bacterial exudate (Figure 4). The dieback caused the leaves to wilt and gradually turn brown, with partial (10 to 90%) and total (100%) coverage (Figure 5).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePerformance of automatic detection models for major diseases.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTraining identified three model variants named V2-3M, V3-3M and V4-3M. These models were evaluated based on the performance of standard metrics, namely precision, recall and Mean Average Precision at an IoU of 50 % (mAP@50) (Figure 6). The evaluation reveals contrasting performances in the detection of symptoms of anthracnose, bacterial disease and mango tree decline. The V2-3M model stands out with its best results, with a mAP@50 of 50,3 %, precision of 55,1 % and recall of 49,1 %. The V4-3M model performs slightly worse, with an mAP@50 of 46.8%, an accuracy of 52 % and a recall of 45,3 %. In contrast, the V3-3M model shows the lowest results, with an mAP@50 of 23,5 % and a recall of only 26,7 %, despite relatively good precision (56,8 %).\u0026nbsp;\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003e\u003cstrong\u003eDistinctive symptoms of mango tree diseases: a solid basis for establishing an automatic diagnostic system.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAccurate identification of phytopathological symptoms is a fundamental prerequisite for the development of digital diagnostic tools, particularly those using AI. The symptoms documented in orchards largely corroborate the classic descriptions established by researchers. Indeed, for anthracnose, observations of irregular brown spots on leaves, sunken black lesions on fruit and necrosis of young twigs are consistent with the work of\u0026nbsp;\u003csup\u003e17\u003c/sup\u003e and\u0026nbsp;\u003csup\u003e18\u003c/sup\u003e, which highlight similar symptoms of anthracnose in humid mango-producing areas, where relative humidity and poor aeration promote infection. The work of\u0026nbsp;\u003csup\u003e19\u003c/sup\u003ehas also demonstrated the early appearance of symptoms as early as the flowering stage, directly influencing yield.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;With regard to bacterial disease, the photographed symptoms captured, such as translucent oily spots, leaf perforations and exudates on fruit, are typical of infection by Xanthomonas campestris pv. mangiferaeindicae. These symptoms were described by\u0026nbsp;\u003csup\u003e6\u003c/sup\u003e in Burkina Faso and West Africa, who highlighted the rapid progression of bacterial lesions. Recognition of these visual characteristics is crucial in order to differentiate this disease from anthracnose, whose lesions can sometimes overlap on the fruit. With regard to decline, symptomatic analysis reveals leaf wilting, progressive browning of twigs, and fruit abortion\u0026nbsp;\u003csup\u003e20\u003c/sup\u003e. This description corresponds to the symptoms associated with complex fungal infections, particularly \u003cem\u003eLasiodiplodia spp\u003c/em\u003e, as reported by\u0026nbsp;\u003csup\u003e21\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll of this symptomatic data thus constitutes a consistent database for training AI models, particularly through deep learning, to identify major diseases. This has been demonstrated in the work of\u0026nbsp;\u003csup\u003e22\u003c/sup\u003e on the automatic identification of wheat diseases in Europe using AI algorithms. Also, based on the training of approximately 8,438 images of symptoms acquired on PlantVillage,\u0026nbsp;\u003csup\u003e23\u003c/sup\u003e have set up a system for the automatic detection of diseases in vineyards and mango trees in India. This will constitute a database of the symptoms of the main diseases affecting mango trees and, above all, a significant step forward in overcoming the lack of phytosanitary expertise and speeding up decision-making in the field by producers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBest model for automatic detection of major mango tree diseases.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe performance achieved by the V2-3M, V3-3M and V4-3M models demonstrates differentiated capabilities in detecting the symptoms of major mango tree diseases from annotated images with accuracies greater than 50 %. The V2-3M model, with an mAP@50 of 50.3%, appears to be the most effective, confirming the importance of good parameterisation and optimised architecture in automatic detection tasks. Its performance combines relatively balanced precision (55,1 %) and recall (49,1 %), indicating that it can be used as a starting point for a functional prototype of a mobile plant health diagnosis tool\u0026nbsp;\u003csup\u003e24\u003c/sup\u003e. These metrics show a good compromise between the model's ability to correctly locate symptoms and its ability to minimise both false positives and false negatives.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOn the other hand, the poor performance of the V3-3M model (mAP@50 = 23,5 % ; recall = 26,7 %) despite similar accuracy (56,8 %) highlights limitations in the generalisation of learning. This model seems to struggle to capture symptomatic diversity, particularly due to reduced sensitivity to morphological and chromatic variations in symptoms.\u003c/p\u003e\n\u003cp\u003eThe metrics of the V4-3M model, which are intermediate (mAP@50 = 46,8 %) and reflect a certain robustness, albeit less marked than V2-3M, show that improved performance depends on a compromise between architectural complexity, annotation quality and training image diversity. This finding is consistent with the observations of\u0026nbsp;\u003csup\u003e25\u003c/sup\u003e, who showed that training models on highly diversified image databases significantly improves the robustness of artificial intelligence tools applied to agriculture.\u003c/p\u003e\n\u003cp\u003eFurthermore, the satisfactory performance of the V2-3M model is in line with recent research into leaf disease detection using deep learning. For example,\u0026nbsp;\u003csup\u003e26\u003c/sup\u003e achieved a mAP of 51% on tomato diseases using CNN architectures.\u0026nbsp;\u003csup\u003e12\u003c/sup\u003e achieved 99 % accuracy for a cashew anthracnose diagnosis model using tools such as Mobilenet and Inception. In terms of performance, the V2-3M model is the most effective. Its ability to balance accuracy (55,1 %) and recall (49,1 %), combined with an mAP@50 of 50,3 %, makes it the most suitable for initial operational application. The V4-3M model, although fairly close, remains slightly less robust, while V3-3M should be ruled out due to its insufficient results. Thus, V2-3M provides a solid basis for an automated diagnostic system, with prospects for improvement.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThese results confirm the feasibility of using artificial intelligence models for the automatic detection of major mango tree diseases, while highlighting the need to continue optimising algorithms and enriching training databases\u003csup\u003e27\u003c/sup\u003e. The integration of such a tool into a mobile application could strengthen early detection capabilities and improve the responsiveness of phytosanitary interventions in Burkina Faso.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eAnthracnose, bacterial disease and dieback are major phytosanitary constraints affecting mango trees. Unfortunately, producers have little knowledge of the epidemiological aspects of these diseases or the control techniques available. This study was conducted with the aim of designing an accurate automatic diagnostic model for the main diseases affecting mango trees using artificial intelligence. Training the collection of symptomatic images showed that the V2-3M model performed better in terms of metrics for automatic disease detection.\u003c/p\u003e\u003cp\u003eThese relevant results open up new prospects for integrating AI tools into agroecological practices, particularly for phytosanitary monitoring and management in orchards. It would be wise to integrate treatment recommendations into the model, deploy it on a mobile application, and promote it for the benefit of producers, all of which will contribute to effective disease management in mango orchards.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFUNDING\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work benefited from the facilities of ACTS/RIMP/FONRID/INERA-01/2023 fund\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research was conducted at the Regional Center of Excellence in Fruits and Vegetables (CRE/FL) based at the Institute of Environment and Agronomic Research (INERA) in Farako-bâ, Burkina Faso.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDZO\u0026nbsp;\u003c/strong\u003e: Conceptualization; Funding acquisition; Methodology; Resources; Roles/Writing - original draft,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eKE :\u0026nbsp;\u003c/strong\u003eManuscript’s writing\u0026nbsp;,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDZO, KE, ZI, KZCO :\u0026nbsp;\u003c/strong\u003eData collection,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eKE\u003c/strong\u003e and\u0026nbsp;\u003cstrong\u003eKZCO\u003c/strong\u003e : Initial processing, data treatement and analysis,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eZI, KZ, LB, MS, WI\u003c/strong\u003e : Critical revision and improvement of the scientific content\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll authors have read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement (mandatory)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData from this study will be made available to the journal as needed via the corresponding author's e-mail address.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors certify that no competing interests or conflicts of interest, whether institutional, financial or personal, have interfered with the scientific integrity, methodology used or objectivity of the results presented in this article.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e\u003cem\u003eFood and agriculture\u003c/em\u003e organization (FAO)\u003cem\u003e- Statistical Yearbook\u003c/em\u003e. \u003cem\u003eFAO\u003c/em\u003e Rome, (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAPEX. \u003cem\u003eFiche Sectorielle: Mangue Du Burkina Faso\u003c/em\u003e. (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.apexb.bf/catalogue-de-produits/mangue?\u003c/span\u003e\u003cspan address=\"https://www.apexb.bf/catalogue-de-produits/mangue?\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e doi:https://www.apexb.bf/catalogue-de-produits/mangue?\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMinistry of Agriculture and Halieutic Resources (MARAH). Fili\u0026egrave;re Mangue: Les Acteurs Font Le Bilan de La Campagne.. (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZida, I. et al. Effectiveness of Four Integrated Pest Management Approaches in the Control of Fruit Flies (Diptera: Tephritidae) in Mango Agro-Ecosystems in the South-Sudanian Zone of Burkina Faso. \u003cem\u003eAdv. Ento- mology\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 124\u0026ndash;142 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBougoum, H. et al. Caract\u0026eacute;risation de Colletotrichum spp agent pathog\u0026egrave;ne de l \u0026rsquo; anthracnose du manguier (Mangifera indica L.) au Burkina Faso. \u003cem\u003eJ. Appl. Biosci.\u003c/em\u003e \u003cb\u003e190\u003c/b\u003e, 19999\u0026ndash;20020 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZombr\u0026eacute;, C. \u003cem\u003ediversite genetique et pathologique de xanthomonas citripv. mangiferaeindicae, bacterie responsable de la maladie des taches noires du manguier au burkina faso et en afrique de l\u0026rsquo;ouest \u0026raquo;\u003c/em\u003e (Universit\u0026eacute; Joseph Ki-Zerbo, Ouagadodugou, 2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDianda, Z. O. et al. Pr\u0026eacute;valence du dess\u0026egrave;chement du manguier et evaluation de la fr\u0026eacute;quence des champignons associ\u0026eacute;s \u0026agrave; la maladie au Burkina Faso. \u003cem\u003eJ. Appl. Biosci.\u003c/em\u003e \u003cb\u003e126\u003c/b\u003e, 12686 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDembele, D. D. \u0026amp; Camara, B. Evaluation pr\u0026eacute; et post-r\u0026eacute;colte de l \u0026rsquo; incidence et de la s\u0026eacute;v\u0026eacute;rit\u0026eacute; de l \u0026rsquo; anthracnose du manguier au nord de la C\u0026ocirc;te d \u0026rsquo; Ivoire. \u003cem\u003eResearchGate\u003c/em\u003e 1\u0026ndash;2 doi: (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.researchgate.net/publication/341314189\u003c/span\u003e\u003cspan address=\"https://www.researchgate.net/publication/341314189\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePaul-alfred, K. K. Profitability and Competitiveness of the Mango Sector in Ivory Coast. \u003cem\u003eInt. J. Econ. Financ Res.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 12\u0026ndash;16 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePIP COLEACP. Nouveaux ravageurs et maladies invasives. Bact\u0026eacute;riose du manguier provoqu\u0026eacute;e par Xanthomonas citri pv. mangiferaeindicae. \u003cem\u003eCOLEACP PIP\u003c/em\u003e 1\u0026ndash;14 p. doi: (2012). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.coleacp.org/pip\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLahrache, R. L \u0026rsquo; impact de l \u0026rsquo; intelligence artificielle sur la prise de d\u0026eacute;cision The Impact of Artificial Intelligence in Decision Making. 7, 660\u0026ndash;678 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJayaprakash, V. \u0026amp; Rajagopal, M. K. cashew dataset generation using augmentation and ralsgan and a transfer learning based tinyml approach abstract: \u003cem\u003earXiv Prepr. arXiv2304.08766\u003c/em\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTirkey, D., Singh, K. K. \u0026amp; Tripathi, S. Performance analysis of AI-based solutions for crop disease identification, detection, and classification. \u003cem\u003eSmart Agric. Technol.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 100238 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDianda, O. et al. Pr\u0026eacute;valence du dess\u0026egrave;chement du manguier et evaluation de la fr\u0026eacute;quence des champignons associ\u0026eacute;s \u0026agrave; la maladie au Burkina Faso. \u003cem\u003eJ. Appl. Biosci.\u003c/em\u003e 12686\u0026ndash;12699. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dx.doi.org/10.4314/jab.v126i1.6\u003c/span\u003e\u003cspan address=\"10.4314/jab.v126i1.6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eINSD. Profil de Pauvret\u0026eacute; de La R\u0026eacute;gion Des Hauts Bassins. (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eINSD. \u003cem\u003e5\u0026egrave; Recensement G\u0026eacute;n\u0026eacute;ral de La Population et de l\u0026rsquo;Habitat\u003c/em\u003e. (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.insd.bf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e doi:www.insd.bf\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDembele, D. D., Og, L., Georges, D. \u0026amp; Amari, E. Pre and postharvest assessment of mango anthracnose incidence and severity in the north of C\u0026ocirc;te d \u0026rsquo; Ivoire Pre and postharvest assessment of mango anthracnose incidence and severity in the north of C\u0026ocirc;te d \u0026rsquo; Ivoire. 14 p. (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4314/ijbcs.v13i6.24\u003c/span\u003e\u003cspan address=\"10.4314/ijbcs.v13i6.24\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDofuor, A. K. et al. Mango anthracnose disease: the current situation and direction for future research. \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1\u0026ndash;18 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKankam, F. \u0026amp; Adomako, J. Anthracnose Disease of Mango: Epidemiology, Impact and Management Options. (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5772/intechopen.105934\u003c/span\u003e\u003cspan address=\"10.5772/intechopen.105934\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDianda, Z. O. et al. P. Pr\u0026eacute;valence du dess\u0026egrave;chement du manguier et evaluation de fr\u0026eacute;quence des champignons associ\u0026eacute;s \u0026agrave; la maladie au Burkina Faso. \u003cem\u003eJ. Appl. Biosci.\u003c/em\u003e 12686\u0026ndash;12699. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dx.doi.org/10.4314/jab.v126i1.6\u003c/span\u003e\u003cspan address=\"10.4314/jab.v126i1.6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOumarou Zo\u0026eacute;yand\u0026eacute; Dianda1, Issa Wonni1, L\u0026eacute;onard Ou\u0026eacute;draogo1, Philippe Sankara2, Charlotte 4 Tollenaere1, 3, Emerson M. Del Ponte4, D. F. Lasiodiplodia species associated with mango (\u003cem\u003eMangifera indica\u003c/em\u003e L.) decline in Burkina Faso and influence of climatic factors on the disease distribution. (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePicon, A. \u0026amp; Alvarez-Gila, A. C, M. S. \u0026amp; Amaia Ortiz-Barredo, Jone Echazarra, A. J. R\u0026eacute;seaux neuronaux convolutifs profonds pour la classification des maladies des cultures \u0026agrave; l\u0026rsquo;\u0026eacute;tat sauvage \u0026agrave; partir d\u0026rsquo;appareils de capture mobiles. \u003cem\u003eScienceDirect\u003c/em\u003e 161, 280\u0026ndash;290 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRao, U. S., Swathi, R., Sanjana, V., Arpitha, L. \u0026amp; Chandrasekhar, K. Deep Learning Precision Farming: Grapes and Mango Leaf Disease Detection by Transfer Learning. \u003cem\u003eGlob. Transitions Proc.\u003c/em\u003e 2, 535-544p (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eElkadi, K. Exp\u0026eacute;rimentation d\u0026rsquo;un mod\u0026egrave;le de d\u0026eacute;tection pr\u0026eacute;coce des maladies de la tomate par apprentissage profond. \u003cem\u003eRev. Marocaine Prot. des. Plantes\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, 19\u0026ndash;30 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamcharan, A., Mccloskey, P., Baranowski, K. \u0026amp; Mbilinyi, N. A. Mobile-Based Deep Learning Model for Cassava Disease Diagnosis. \u003cem\u003eFront. Plant. Sci.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 1\u0026ndash;8 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoulent, J., Foucher, S. \u0026amp; Th\u0026eacute;au, J. Convolutional Neural Networks for the Automatic Identification of Plant Diseases. \u003cem\u003eFront. Plant. Sci.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 15 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHossain, M. A., Sakib, S., Abdullah, H. M. \u0026amp; Arman, S. E. Deep learning for mango leaf disease identification: A vision transformer perspective. \u003cem\u003eHeliyon\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 36361 (2024).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"mango tree, diseases, artificial intelligence, automatic diagnosis, Burkina Faso","lastPublishedDoi":"10.21203/rs.3.rs-7829136/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7829136/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn Burkina Faso, the mango industry faces numerous phytosanitary constraints. This situation is exacerbated by producers' lack of knowledge about diseases and their symptoms, and the low uptake of control technologies. This study aims to develop a tool using artificial intelligence-based approaches for better integrated management of the main diseases affecting mango trees.\u003c/p\u003e\u003cp\u003eSurveys conducted in 40 mango orchards yielded 11,001 images of the characteristic symptoms of three major diseases, including 4,527 for anthracnose, 3,038 for bacterial disease and 3,436 for dieback. The data underwent several preprocessing steps. To ensure a balance between disease classes, 3,000 images per class were selected, for a total of 9,000 images. The images were then annotated and used for training with the pre-trained YOLOv11 model. Following training, three models were formed : V2-3M (MAP@50\u0026thinsp;=\u0026thinsp;50.3% ; precision\u0026thinsp;=\u0026thinsp;55.1% ; recall\u0026thinsp;=\u0026thinsp;49.1%), V3-3M (MAP@50\u0026thinsp;=\u0026thinsp;23% ; precision\u0026thinsp;=\u0026thinsp;56.8% ; recall\u0026thinsp;=\u0026thinsp;26.7%), and V4-3M (MAP@50\u0026thinsp;=\u0026thinsp;46.8% ; precision\u0026thinsp;=\u0026thinsp;52% ; recall\u0026thinsp;=\u0026thinsp;45.3%). The most effective model is the V2-3M, as it combines 55.1% accuracy, 50.3% MAP@50 and 49.1% recall.\u003c/p\u003e\u003cp\u003eIn order to improve diagnosis and phytosanitary management in orchards, it would be useful to integrate treatment recommendations into the model, then deploy it on a mobile application and promote it to producers.\u003c/p\u003e","manuscriptTitle":"Evaluation of state-of-the-art models for improving the diagnosis of mango tree diseases through supervised learning of symptom images in Burkina Faso","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-13 07:59:20","doi":"10.21203/rs.3.rs-7829136/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-11T05:15:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"99307988563381004942591408318993275720","date":"2026-04-20T12:22:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"75555148691676737383067153822348684187","date":"2026-03-23T16:36:41+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-13T06:30:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"281596109643282957064993017059991404176","date":"2026-01-19T12:04:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"296770740346100154457008044908190897148","date":"2026-01-18T10:34:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"110523610180534089449693048545077046290","date":"2026-01-17T06:29:06+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-02T14:22:21+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-13T19:25:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-13T03:05:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-13T03:05:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-10-10T16:40:25+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ee066dd7-112b-4831-949b-4d3854ae1054","owner":[],"postedDate":"November 13th, 2025","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-11T05:15:06+00:00","index":164,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":57293873,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":57293874,"name":"Health sciences/Diseases"},{"id":57293875,"name":"Health sciences/Health care"}],"tags":[],"updatedAt":"2025-11-13T07:59:20+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-13 07:59:20","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7829136","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7829136","identity":"rs-7829136","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}