Partial concordance of vascular phenotypes across retinal and peripheral microvascular beds in the OphtalmoLaus cohort

Partial concordance of vascular phenotypes across retinal and peripheral microvascular beds in the OphtalmoLaus cohort | 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 Partial concordance of vascular phenotypes across retinal and peripheral microvascular beds in the OphtalmoLaus cohort Daniel Bach, Sophie Strebel, Adham Elwakil, Sandro De Zanet, The VascX Research Consortium, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9243114/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 Background: Whether vascular phenotypes measured in one anatomical territory reflect those observed in other vascular beds remains unclear. Retinal imaging is increasingly used as a source of systemic vascular biomarkers, yet direct comparisons with oral and peripheral microcirculatory phenotypes within the same individuals remain limited. Methods: We investigated the cross-territory organisation of multimodal vascular phenotypes in a deeply phenotyped subset of the OphtalmoLaus cohort. The analysis included 2,523 participants with ocular imaging available after quality control, of whom 82 also underwent sublingual and sublabial microcirculatory imaging. Vascular features were derived from structural optical coherence tomography (OCT), OCT angiography (OCTA), colour fundus imaging (CFI), sublingual imaging, sublabial imaging, and an exploratory family of capillary network-derived OCTA features. Pairwise Pearson correlation coefficients were computed across all available feature pairs to examine within-modality structure and cross-modality concordance. Results: The global correlation matrix showed a clear block-diagonal organisation, with stronger within-modality than between-modality correlations. Distinct internal clustering was observed across OCT, OCTA, CFI, and exploratory capillary network-derived feature families, indicating coherent modality-specific vascular phenotype domains. Cross-modality correlations were generally weaker and more heterogeneous, arguing against a single strongly coupled vascular phenotype shared across all territories. Among extra-ocular comparisons, sublingual and sublabial features showed the clearest cross-bed concordance, whereas retinal structural and perfusion-related phenotypes showed only limited correspondence with oral and peripheral microcirculatory measures. Within the retinal domain, OCTA and CFI demonstrated the strongest cross-modality alignment, consistent with partial overlap between perfusion- and fundus-derived vascular phenotypes. Conclusions: Multimodal vascular phenotyping in OphtalmoLaus reveals a structured but only partially shared organisation of vascular traits across anatomical territories. These findings support a model of selective, territory-specific coupling across vascular beds and suggest that retinal and peripheral microcirculatory imaging provide overlapping yet non-redundant information. Biomedical Engineering Microcirculation OCT angiography retinal biomarkers multimodal imaging vascular phenotyping correlation analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction The microcirculation plays a central role in tissue oxygenation, nutrient delivery, and vascular homeostasis, and its disruption is increasingly recognised as an important component of cardiovascular, metabolic, and neurovascular disease. Because microvascular dysfunction cannot be assessed easily in most organs in vivo, there is growing interest in non-invasive imaging approaches that provide access to vascular structure and perfusion across different anatomical territories [ 1 , 2 ]. Among available imaging sites, the retina has attracted particular attention as a readily accessible vascular bed in which both large-vessel geometry and capillary-level organisation can be quantified non-invasively. Retinal vascular calibre, tortuosity, and other microvascular signs have been associated with hypertension, diabetes, systemic inflammation, stroke, coronary heart disease, and long-term cardiovascular risk, supporting the idea that retinal imaging may provide a window into systemic vascular health [ 3 – 6 ]. Recent advances in ocular imaging have further expanded this field. Structural optical coherence tomography (OCT) enables detailed measurement of retinal and choroidal morphology, while OCT angiography (OCTA) provides depth-resolved assessment of retinal perfusion and capillary network organisation. OCTA abnormalities have been reported across a range of systemic conditions, including diabetes, hypertension, cardiovascular disease, and neurodegenerative disorders, reinforcing the promise of ocular imaging as a source of vascular biomarkers. At the same time, recent reviews have emphasised that interpretation remains complex, and that retinal vascular phenotypes likely reflect only part of the broader systemic microcirculatory landscape [ 7 ]. Outside the eye, sublingual microcirculation imaging has emerged as another important in vivo approach for direct assessment of human microvascular function and architecture. Consensus recommendations have standardised the acquisition and interpretation of sublingual microcirculatory measurements, particularly in critical care and haemodynamic research, and the sublingual bed is increasingly used as a practical surrogate for systemic microvascular status. However, even in that setting, microcirculatory coherence across vascular territories cannot be assumed [ 1 ]. A key unresolved question is therefore whether vascular phenotypes measured in one anatomical territory meaningfully reflect those observed in another. The retina is often described as a “window” to systemic microcirculation, yet direct comparisons between retinal, oral, and peripheral vascular beds within the same individuals remain limited. Most studies have focused on a single modality or a single vascular territory, making it difficult to determine whether observed vascular traits represent shared systemic biology, territory-specific regulation, or a combination of both [ 2 , 7 , 8 ]. Deeply phenotyped multimodal cohorts now provide an opportunity to address this question more directly. By combining structural OCT, OCTA, colour fundus-derived vascular phenotypes, and peripheral microcirculatory imaging in the same participants, it becomes possible to examine the extent to which vascular features cluster within modalities, align across related vascular beds, or remain largely compartment-specific. Such analyses are important not only for biological understanding, but also for the interpretation of retinal biomarkers in cardiovascular and systemic disease research. In the present work, we investigated the cross-territory organisation of vascular phenotypes derived from ocular and extra-ocular imaging modalities within the same individuals. We hypothesised that correlations would be stronger within modalities than across modalities, and that anatomically related peripheral vascular beds would show greater concordance than retina–periphery comparisons. More broadly, we sought to determine whether multimodal vascular phenotypes support a model of global vascular coupling or, instead, a more selective and territory-specific organisation of the human microvasculature. 2. Methods 2.1 Study design and population This analysis was performed within OphtalmoLaus, an ophthalmic ancillary study nested in the population-based CoLaus|PsyCoLaus cohort in Lausanne, Switzerland. Participants underwent standardised ophthalmic imaging at the Jules Gonin Eye Hospital between November 2015 and November 2025. A subgroup also underwent non-invasive imaging of the sublingual mucosa and inner lower lip microcirculation using incident dark-field imaging. 2.2 Imaging modalities and feature extraction The analysis integrated vascular phenotypes derived from multiple ocular and extra-ocular imaging modalities, including structural optical coherence tomography (OCT), OCT angiography (OCTA), colour fundus imaging (CFI), sublingual imaging, sublabial imaging, and an exploratory family of capillary network-derived OCTA features. All ocular image processing and feature extraction were performed within the Cohort Builder pipeline [ 9 ], using modality-specific analysis modules. Structural OCT features were extracted using Discovery (retinaAI) [ 10 ], OCTA features using CapX [ 11 ], and colour fundus image-derived vascular features using VascX [ 12 ]. A detailed description of ocular, sublingual, and sublabial phenotypes, together with the full list of extracted vascular features and their technical definitions, has been reported previously [ 13 ]. In addition, capillary-derived network-level features were obtained from preliminary outputs of an extended version of the CapX pipeline currently under development, designed to provide a more detailed characterisation of microvascular organisation in superficial and deep plexus OCTA en face images. Overall, the integrated feature set was designed to represent vascular phenotypes across multiple anatomical territories and spatial scales, enabling assessment of concordance and complementarity between retinal and peripheral microvascular domains. 2.3 Statistical analysis Pairwise Pearson correlation coefficients were computed across all available feature pairs and visualised as heatmaps. For each pairwise correlation, all participants with non-missing data for both features were included, such that the effective sample size could vary across modality pairs depending on data availability after quality control. Correlation matrices were generated for each pair of feature families as well as for the full integrated feature set. Features were grouped by modality to facilitate interpretation of within-modality structure and cross-modality associations. Given the exploratory aim of this initial analysis, correlation patterns were interpreted qualitatively from the resulting matrices, with emphasis on the relative organisation of within- and between-modality associations rather than formal hypothesis testing. 3. Results 3.1 Analytic sample and feature overview The analysis included 2,523 participants with multimodal vascular phenotyping available after quality control. Availability varied across imaging modalities, with 82 participants contributing sublingual and sublabial microcirculatory imaging, while ocular imaging modalities were available in substantially larger subsets (structural OCT: 2,085 participants; OCTA: 1,788 participants; colour fundus imaging: 2,460 participants). Pairwise correlations were therefore computed using all participants with non-missing data for each feature pair, such that the effective sample size varied across modality combinations. The integrated feature set comprised 29 structural OCT features, 34 OCTA features, 49 colour fundus image-derived features, 6 sublingual features, 6 sublabial features, and 29 exploratory capillary network-derived features. 3.2 Cross-territory organisation of vascular phenotypes We examined the correlation structure of vascular features across imaging modalities and anatomical territories (Figs. 1 – 4 ; Supplementary Figs. S1–S12). The global correlation matrix revealed a clear block-diagonal organisation, with stronger and more coherent within-modality than between-modality correlations. This was confirmed quantitatively, with a higher median absolute correlation within modalities than across modalities (median |r| = 0.12 vs 0.05). By contrast, cross-modality relationships were generally weaker and more heterogeneous. Across all between-modality feature pairs, only a small proportion showed moderate associations (|r| ≥ 0.3 in 1.2% of pairs), and strong correlations were rare. These findings argue against the presence of a single strongly coupled vascular phenotype shared across all anatomical territories. Within peripheral microcirculation, sublingual and sublabial features showed the clearest cross-bed concordance outside the retinal domain (Fig. 2 ). The overall strength of association was modest (median |r| = 0.16), but substantially higher than for most cross-modality comparisons. The strongest positive association was observed between homologous average capillary length measures (r = 0.63, n = 70), while inverse relationships were observed between capillary length and segment-based descriptors (minimum r = − 0.57, n = 70), reflecting structural dependencies within the capillary network. Retinal-derived features showed more limited correspondence with peripheral microcirculatory measures. For example, sublingual–OCTA associations were weak overall (median |r| = 0.07), with no correlations exceeding |r| ≥ 0.3. Similar patterns were observed for sublingual–OCT and sublingual–capillary comparisons, indicating sparse and heterogeneous relationships without consistent cross-feature structure. Within the retinal domain, OCT, OCTA, and CFI showed partial but incomplete concordance. Among these, OCTA and CFI demonstrated the strongest cross-modality structure (Fig. 4 ), with moderate associations observed for selected feature pairs, particularly involving vessel tortuosity descriptors (maximum r = 0.47). However, overall correlations remained modest (median |r| = 0.04), indicating that substantial modality-specific information is retained. Across all modality pairs, both positive and negative correlations were observed, suggesting that different feature domains capture distinct and sometimes opposing aspects of vascular organisation. Overall, the correlation structure supports a model of selective, territory-specific coupling across vascular beds, with substantially greater redundancy within modalities than between them. 4. Discussion In this exploratory analysis of multimodal vascular phenotypes within the OphtalmoLaus cohort, we examined how imaging-derived vascular features relate across retinal and peripheral vascular territories. The global correlation structure was dominated by stronger within-modality than between-modality associations, with a higher median absolute correlation within modalities than across modalities (median |r| = 0.12 vs 0.05). This indicates that vascular features are organised into modality-specific domains rather than forming a single, strongly coupled network across anatomical territories. Cross-territory correlations were generally modest and heterogeneous, with only a small proportion of feature pairs showing moderate associations. Retinal structural and perfusion-related features showed limited correspondence with oral and peripheral microcirculatory measures, suggesting that retinal imaging captures only part of the broader systemic vascular phenotype. These findings support a model of selective rather than global coupling across vascular beds. Within peripheral microcirculation, sublingual and sublabial features showed the clearest cross-bed concordance, with moderate associations observed for homologous features such as average capillary length (r = 0.63). At the same time, inverse relationships between capillary length and segment-based descriptors highlight the non-redundant nature of these measures. Given the smaller sample size for oral microcirculatory imaging, these findings should be interpreted cautiously. Within the retina, OCTA and CFI showed the strongest cross-modality alignment, particularly for vascular geometry-related features, although overall correlations remained modest. The incomplete correspondence between OCT, OCTA, and CFI further supports the view that structural, perfusion, and large-vessel phenotypes represent distinct but related aspects of vascular organisation. This study has several limitations. It is cross-sectional and correlation-based, and does not account for systemic covariates such as age, sex, or cardiovascular risk factors. The effective sample size varied across modality pairs, and part of the feature set included exploratory outputs from an extended version of the CapX pipeline. These findings should therefore be interpreted as an initial characterisation of cross-territory vascular organisation rather than definitive evidence of biological coupling. Overall, multimodal vascular phenotyping reveals a structured but only partially shared organisation of vascular traits across anatomical territories. Retinal and peripheral microcirculatory imaging provide overlapping yet non-redundant information, supporting the value of integrated multimodal approaches for studying vascular biology. 5. Conclusion Multimodal vascular phenotyping in OphtalmoLaus reveals a structured but only partially shared organisation of vascular traits across anatomical territories. Correlations were consistently stronger within-modality than between-modality correlations (median |r| = 0.12 vs 0.05), supporting the presence of modality-specific vascular phenotype domains. Cross-territory associations were generally modest, with limited correspondence between retinal and peripheral microcirculatory features. These findings suggest that vascular beds are selectively rather than globally coupled, and that retinal and peripheral imaging provide overlapping yet non-redundant information for the characterisation of microvascular organisation. Abbreviations CFI, colour fundus imaging; CVI, choroidal vascularity index; FAZ, foveal avascular zone; GCL+IPL, ganglion cell layer + inner plexiform layer; INL+OPL, inner nuclear layer + outer plexiform layer; OCT, optical coherence tomography; OCTA, optical coherence tomography angiography; RNFL, retinal nerve fibre layer. Declarations Acknowledgements This work was supported by the Swiss Personalized Health Network (grant 2018DRI13 to Thomas J. Wolfensberger), the Claire and Selma Kattenburg Foundation (Prix Kattenburg 2022 to Mattia Tomasoni, 2023 to Ciara Bergin and Mattia Tomasoni and 2025 to Mattia Tomasoni), and the Swiss National Science Foundation (grant no. CRSII5 209510 to Sven Bergmann). Ethics approval Approved by CER-VD (PB_2019-00168). Written informed consent was obtained from all participants. References Ince C, Boerma EC, Cecconi M, De Backer D, Shapiro NI, Duranteau J et al (2018) Second consensus on the assessment of sublingual microcirculation in critically ill patients: results from a task force of the European Society of Intensive Care Medicine. Intensive Care Med 44:281–299 Yuan Y, Dong M, Wen S, Yuan X, Zhou L (2024) Retinal microcirculation: A window into systemic circulation and metabolic disease. Exp Eye Res 242:109885 Wong TY, Islam FMA, Klein R, Klein BEK, Cotch MF, Castro C et al (2006) Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (MESA). Invest Ophthalmol Vis Sci 47:2341–2350 McGeechan K, Liew G, Macaskill P, Irwig L, Klein R, Klein BEK et al (2009) Meta-analysis: retinal vessel caliber and risk for coronary heart disease. Ann Intern Med 151:404–413 Seidelmann SB, Claggett B, Bravo PE, Gupta A, Farhad H, Klein BE et al (2016) Retinal Vessel Calibers in Predicting Long-Term Cardiovascular Outcomes: The Atherosclerosis Risk in Communities Study. Circulation 134:1328–1338 McGeechan K, Liew G, Macaskill P, Irwig L, Klein R, Klein BEK et al (2009) Prediction of incident stroke events based on retinal vessel caliber: a systematic review and individual-participant meta-analysis. Am J Epidemiol 170:1323–1332 Optical coherence tomography angiography (2024) of the retina and choroid in systemic diseases. Prog Retin Eye Res 103:101292 Retinal Vascular Signs (2011) A Window to the Heart? Revista Española de Cardiología. (English Edition) 64:515–521 Mousavi S, Garjani A, Elwakil A, Brock LP, Dherse A, Forestier E et al (2024) Cohort Builder: A Software Pipeline for Generating Patient Cohorts with Predetermined Baseline Characteristics from Medical Records and Raw Ophthalmic Imaging Data. Stud Health Technol Inform 316. 10.3233/SHTI240613 De Zanet S, Ciller C, Wolf S, Sznitman R, Pathological (2017) OCT Retinal Layer Segmentation Using Branch Residual U-Shape Networks. Medical Image Computing and Computer Assisted Intervention – MICCAI 2017; 294–301 Elwakil A, Martin T, Quiros JV, Liefers B, Kropp M, Duriez A et al CapillaryX: A Fine-Tunable Pipeline for OCTA Segmentation and Feature Extraction. 2025 [cited 24 Mar 2026]. 10.21203/rs.3.rs-8176394/v1 Vargas Quiros J, Liefers B, van Garderen KA, Vermeulen JP, Klaver C (2025) VascX Models: Deep Ensembles for Retinal Vascular Analysis From Color Fundus Images. Transl Vis Sci Technol 14:19 Meloni I, Elwakil A, Aurélia G, Flavie T, Navarro A, Bagatella L et al Cohort profile: OphtalmoLaus, an extension of the CoLaus|PsyCoLaus cohort to investigate the relationships between ocular, cardiovascular, and cognitive parameters. 2025 [cited 24 Mar 2026]. 10.21203/rs.3.rs-7660406/v1 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryMaterial.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-9243114","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":613199262,"identity":"d9a00556-c345-44ad-968e-5caf2071e595","order_by":0,"name":"Daniel Bach","email":"","orcid":"","institution":"Platform for Research in Ocular Imaging, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Bach","suffix":""},{"id":613199263,"identity":"b9ee85bb-642b-40eb-b652-06f8c3bb9640","order_by":1,"name":"Sophie Strebel","email":"","orcid":"","institution":"Platform for Research in Ocular Imaging, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"","lastName":"Strebel","suffix":""},{"id":613199264,"identity":"c7ba44dc-ef70-4586-a212-3b39d10298ac","order_by":2,"name":"Adham Elwakil","email":"","orcid":"","institution":"Platform for Research in Ocular Imaging, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Adham","middleName":"","lastName":"Elwakil","suffix":""},{"id":613199265,"identity":"c0d20904-19ad-47fb-9418-7332b4c5b0eb","order_by":3,"name":"Sandro De Zanet","email":"","orcid":"","institution":"Ikerian AG, Bern, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Sandro","middleName":"","lastName":"De Zanet","suffix":""},{"id":613199266,"identity":"55d567b7-eff4-404f-8907-fdb58db84023","order_by":4,"name":"The VascX Research Consortium","email":"","orcid":"","institution":"The VascX Research Consortium","correspondingAuthor":false,"prefix":"","firstName":"The","middleName":"VascX Research","lastName":"Consortium","suffix":""},{"id":613199267,"identity":"74325da6-0dc5-4c02-be4f-264b8650604f","order_by":5,"name":"OphtalmoLaus Research Consortium","email":"","orcid":"","institution":"OphtalmoLaus Research Consortium","correspondingAuthor":false,"prefix":"","firstName":"OphtalmoLaus","middleName":"Research","lastName":"Consortium","suffix":""},{"id":613199268,"identity":"3e19cc72-465d-484c-89a9-36c695fc7f61","order_by":6,"name":"Thomas J. Wolfensberger","email":"","orcid":"","institution":"Department of Ophthalmology, University of Lausanne, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"J.","lastName":"Wolfensberger","suffix":""},{"id":613199269,"identity":"20e72e95-5790-4bb6-bb0d-a7f3caef42d9","order_by":7,"name":"Reinier Schlingemann","email":"","orcid":"","institution":"Amsterdam UMC, University of Amsterdam, Department of Ophthalmology, Amsterdam, The Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Reinier","middleName":"","lastName":"Schlingemann","suffix":""},{"id":613199270,"identity":"cd750d0f-89c4-424b-960f-cfa080f5be93","order_by":8,"name":"Ciara Bergin","email":"","orcid":"","institution":"Department of Ophthalmology, University of Lausanne, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne, Switzerland","correspondingAuthor":false,"prefix":"","firstName":"Ciara","middleName":"","lastName":"Bergin","suffix":""},{"id":613199271,"identity":"f99a881f-4e34-453b-8cc7-c52c7aa583b0","order_by":9,"name":"Ilenia Meloni","email":"","orcid":"https://orcid.org/0000-0001-9078-0538","institution":"Platform for Research in Ocular Imaging, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne","correspondingAuthor":false,"prefix":"","firstName":"Ilenia","middleName":"","lastName":"Meloni","suffix":""},{"id":613199272,"identity":"af6c4e98-75d5-4c87-8599-fe7fba1d99b7","order_by":10,"name":"Mattia Tomasoni","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9ElEQVRIie3OsYrCQBCA4QkBbVa2kwkBfQIhIgQLH2aDYBrF7qpDUq1NvPqCj2Fj58qBNgFbwSYSsDYIx9mIGzgJFm5ai/2rHdiPGQCd7l1jxRMb1cAUYASq/6YRSIIP0iGiwsoJFAS8UBBHSVrT3TZJPmFC6z/pOVt2/bAqHLguXxM37svDNoDWfOBaUYyjkDBmzGIFETmpADoH5po1jqMVMGEaXEF2qSS3nPiXiyQ+ocdATfZyi8dzMnRsSRjBPpSQtP3tfaEVzYcfVsSxHeIJ1jPlYV6S/f32KLX9xTnjkyah4yy5Ksh/+DyKUqDT6XQ6ZXdBj1CagO9JHQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-8775-2384","institution":"Platform for Research in Ocular Imaging, Fondation Asile des Aveugles, Jules Gonin Eye Hospital, Lausanne","correspondingAuthor":true,"prefix":"","firstName":"Mattia","middleName":"","lastName":"Tomasoni","suffix":""}],"badges":[],"createdAt":"2026-03-27 09:49:08","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9243114/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9243114/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105728927,"identity":"0c88f6b1-1ebc-4811-b69a-fd823e9130af","added_by":"auto","created_at":"2026-03-30 11:13:01","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":117387,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGlobal correlation structure across multimodal vascular phenotypes.\u003c/strong\u003e Pairwise Pearson correlation matrix of all imaging-derived features, grouped by modality: OCT angiography (OCTA), colour fundus imaging-derived features (CFI), sublingual microcirculation, sublabial (sublip) microcirculation, structural optical coherence tomography (OCT), and capillary-derived features. Each cell represents the correlation coefficient between one pair of features across participants. Diagonal blocks represent within-modality correlations, whereas off-diagonal blocks represent cross-modality associations. The matrix shows stronger clustering within modalities than between modalities, consistent with selective rather than global concordance across vascular territories.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/b602cb21c292f5cf655bc89e.png"},{"id":105729261,"identity":"1fb3493a-03ac-417a-97a0-defc68e29e9e","added_by":"auto","created_at":"2026-03-30 11:14:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":338670,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePearson correlation matrix between sublingual and sublabial microcirculatory features. \u003c/strong\u003ePairwise Pearson correlation coefficients between sublingual-derived features (rows) and sublabial (sublip)-derived features (columns). The matrix shows partial concordance between the two oral vascular beds, with localized clusters of positive and negative association rather than uniform overlap across all features.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/d0ff4ee559d3c0eb7c24199b.png"},{"id":105698451,"identity":"87b36530-4ab5-4acc-bf01-f1e24304068e","added_by":"auto","created_at":"2026-03-30 05:04:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":229657,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePearson correlation matrix between sublingual and OCTA-derived features.\u003c/strong\u003e Pairwise Pearson correlation coefficients between sublingual-derived features (rows) and OCTA-derived features (columns). Compared with oral–oral and retina–retina comparisons, the matrix shows sparse and heterogeneous associations, illustrating limited cross-territory correspondence between oral microcirculatory and retinal perfusion-related phenotypes.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/86d59a4e1481fe801360d209.png"},{"id":105698452,"identity":"3e08d549-c850-40a1-b03f-f1ad782fd4e7","added_by":"auto","created_at":"2026-03-30 05:04:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":305588,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePearson correlation matrix between OCTA and CFI-derived features.\u003c/strong\u003e Pairwise Pearson correlation coefficients between OCTA-derived features (rows) and CFI-derived features (columns). The matrix shows the clearest cross-modality structure within the retinal domain, consistent with partial overlap between perfusion- and fundus-derived vascular phenotypes while preserving substantial modality-specific information.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/cbab17d038c65b4a1acf2ca0.png"},{"id":105730147,"identity":"7246df1a-4930-45ad-a864-54c362f563e6","added_by":"auto","created_at":"2026-03-30 11:22:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1329345,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/fed8796b-295b-4d9c-b483-d366c6cd58a6.pdf"},{"id":105698449,"identity":"2d59414c-b48c-448b-b997-67ff04121eef","added_by":"auto","created_at":"2026-03-30 05:04:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3432562,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-9243114/v1/0921dfe333f55207751697ca.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003ePartial concordance of vascular phenotypes across retinal and peripheral microvascular beds in the OphtalmoLaus cohort\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe microcirculation plays a central role in tissue oxygenation, nutrient delivery, and vascular homeostasis, and its disruption is increasingly recognised as an important component of cardiovascular, metabolic, and neurovascular disease. Because microvascular dysfunction cannot be assessed easily in most organs in vivo, there is growing interest in non-invasive imaging approaches that provide access to vascular structure and perfusion across different anatomical territories [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAmong available imaging sites, the retina has attracted particular attention as a readily accessible vascular bed in which both large-vessel geometry and capillary-level organisation can be quantified non-invasively. Retinal vascular calibre, tortuosity, and other microvascular signs have been associated with hypertension, diabetes, systemic inflammation, stroke, coronary heart disease, and long-term cardiovascular risk, supporting the idea that retinal imaging may provide a window into systemic vascular health [\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRecent advances in ocular imaging have further expanded this field. Structural optical coherence tomography (OCT) enables detailed measurement of retinal and choroidal morphology, while OCT angiography (OCTA) provides depth-resolved assessment of retinal perfusion and capillary network organisation. OCTA abnormalities have been reported across a range of systemic conditions, including diabetes, hypertension, cardiovascular disease, and neurodegenerative disorders, reinforcing the promise of ocular imaging as a source of vascular biomarkers. At the same time, recent reviews have emphasised that interpretation remains complex, and that retinal vascular phenotypes likely reflect only part of the broader systemic microcirculatory landscape [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOutside the eye, sublingual microcirculation imaging has emerged as another important in vivo approach for direct assessment of human microvascular function and architecture. Consensus recommendations have standardised the acquisition and interpretation of sublingual microcirculatory measurements, particularly in critical care and haemodynamic research, and the sublingual bed is increasingly used as a practical surrogate for systemic microvascular status. However, even in that setting, microcirculatory coherence across vascular territories cannot be assumed [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA key unresolved question is therefore whether vascular phenotypes measured in one anatomical territory meaningfully reflect those observed in another. The retina is often described as a \u0026ldquo;window\u0026rdquo; to systemic microcirculation, yet direct comparisons between retinal, oral, and peripheral vascular beds within the same individuals remain limited. Most studies have focused on a single modality or a single vascular territory, making it difficult to determine whether observed vascular traits represent shared systemic biology, territory-specific regulation, or a combination of both [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDeeply phenotyped multimodal cohorts now provide an opportunity to address this question more directly. By combining structural OCT, OCTA, colour fundus-derived vascular phenotypes, and peripheral microcirculatory imaging in the same participants, it becomes possible to examine the extent to which vascular features cluster within modalities, align across related vascular beds, or remain largely compartment-specific. Such analyses are important not only for biological understanding, but also for the interpretation of retinal biomarkers in cardiovascular and systemic disease research.\u003c/p\u003e \u003cp\u003eIn the present work, we investigated the cross-territory organisation of vascular phenotypes derived from ocular and extra-ocular imaging modalities within the same individuals. We hypothesised that correlations would be stronger within modalities than across modalities, and that anatomically related peripheral vascular beds would show greater concordance than retina\u0026ndash;periphery comparisons. More broadly, we sought to determine whether multimodal vascular phenotypes support a model of global vascular coupling or, instead, a more selective and territory-specific organisation of the human microvasculature.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study design and population\u003c/h2\u003e \u003cp\u003eThis analysis was performed within OphtalmoLaus, an ophthalmic ancillary study nested in the population-based CoLaus|PsyCoLaus cohort in Lausanne, Switzerland. Participants underwent standardised ophthalmic imaging at the Jules Gonin Eye Hospital between November 2015 and November 2025. A subgroup also underwent non-invasive imaging of the sublingual mucosa and inner lower lip microcirculation using incident dark-field imaging.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Imaging modalities and feature extraction\u003c/h2\u003e \u003cp\u003eThe analysis integrated vascular phenotypes derived from multiple ocular and extra-ocular imaging modalities, including structural optical coherence tomography (OCT), OCT angiography (OCTA), colour fundus imaging (CFI), sublingual imaging, sublabial imaging, and an exploratory family of capillary network-derived OCTA features.\u003c/p\u003e \u003cp\u003eAll ocular image processing and feature extraction were performed within the Cohort Builder pipeline [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], using modality-specific analysis modules. Structural OCT features were extracted using Discovery (retinaAI) [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], OCTA features using CapX [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], and colour fundus image-derived vascular features using VascX [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. A detailed description of ocular, sublingual, and sublabial phenotypes, together with the full list of extracted vascular features and their technical definitions, has been reported previously [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e In addition, capillary-derived network-level features were obtained from preliminary outputs of an extended version of the CapX pipeline currently under development, designed to provide a more detailed characterisation of microvascular organisation in superficial and deep plexus OCTA en face images.\u003c/p\u003e \u003cp\u003eOverall, the integrated feature set was designed to represent vascular phenotypes across multiple anatomical territories and spatial scales, enabling assessment of concordance and complementarity between retinal and peripheral microvascular domains.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Statistical analysis\u003c/h2\u003e \u003cp\u003ePairwise Pearson correlation coefficients were computed across all available feature pairs and visualised as heatmaps. For each pairwise correlation, all participants with non-missing data for both features were included, such that the effective sample size could vary across modality pairs depending on data availability after quality control. Correlation matrices were generated for each pair of feature families as well as for the full integrated feature set. Features were grouped by modality to facilitate interpretation of within-modality structure and cross-modality associations. Given the exploratory aim of this initial analysis, correlation patterns were interpreted qualitatively from the resulting matrices, with emphasis on the relative organisation of within- and between-modality associations rather than formal hypothesis testing.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Analytic sample and feature overview\u003c/h2\u003e \u003cp\u003eThe analysis included 2,523 participants with multimodal vascular phenotyping available after quality control. Availability varied across imaging modalities, with 82 participants contributing sublingual and sublabial microcirculatory imaging, while ocular imaging modalities were available in substantially larger subsets (structural OCT: 2,085 participants; OCTA: 1,788 participants; colour fundus imaging: 2,460 participants). Pairwise correlations were therefore computed using all participants with non-missing data for each feature pair, such that the effective sample size varied across modality combinations.\u003c/p\u003e \u003cp\u003eThe integrated feature set comprised 29 structural OCT features, 34 OCTA features, 49 colour fundus image-derived features, 6 sublingual features, 6 sublabial features, and 29 exploratory capillary network-derived features.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Cross-territory organisation of vascular phenotypes\u003c/h2\u003e \u003cp\u003eWe examined the correlation structure of vascular features across imaging modalities and anatomical territories (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Supplementary Figs. S1\u0026ndash;S12). The global correlation matrix revealed a clear block-diagonal organisation, with stronger and more coherent within-modality than between-modality correlations. This was confirmed quantitatively, with a higher median absolute correlation within modalities than across modalities (median |r| = 0.12 vs 0.05).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBy contrast, cross-modality relationships were generally weaker and more heterogeneous. Across all between-modality feature pairs, only a small proportion showed moderate associations (|r| \u0026ge; 0.3 in 1.2% of pairs), and strong correlations were rare. These findings argue against the presence of a single strongly coupled vascular phenotype shared across all anatomical territories.\u003c/p\u003e \u003cp\u003eWithin peripheral microcirculation, sublingual and sublabial features showed the clearest cross-bed concordance outside the retinal domain (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The overall strength of association was modest (median |r| = 0.16), but substantially higher than for most cross-modality comparisons. The strongest positive association was observed between homologous average capillary length measures (r\u0026thinsp;=\u0026thinsp;0.63, n\u0026thinsp;=\u0026thinsp;70), while inverse relationships were observed between capillary length and segment-based descriptors (minimum r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.57, n\u0026thinsp;=\u0026thinsp;70), reflecting structural dependencies within the capillary network.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eRetinal-derived features showed more limited correspondence with peripheral microcirculatory measures. For example, sublingual\u0026ndash;OCTA associations were weak overall (median |r| = 0.07), with no correlations exceeding |r| \u0026ge; 0.3. Similar patterns were observed for sublingual\u0026ndash;OCT and sublingual\u0026ndash;capillary comparisons, indicating sparse and heterogeneous relationships without consistent cross-feature structure.\u003c/p\u003e \u003cp\u003eWithin the retinal domain, OCT, OCTA, and CFI showed partial but incomplete concordance. Among these, OCTA and CFI demonstrated the strongest cross-modality structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e), with moderate associations observed for selected feature pairs, particularly involving vessel tortuosity descriptors (maximum r\u0026thinsp;=\u0026thinsp;0.47). However, overall correlations remained modest (median |r| = 0.04), indicating that substantial modality-specific information is retained.\u003c/p\u003e \u003cp\u003eAcross all modality pairs, both positive and negative correlations were observed, suggesting that different feature domains capture distinct and sometimes opposing aspects of vascular organisation. Overall, the correlation structure supports a model of selective, territory-specific coupling across vascular beds, with substantially greater redundancy within modalities than between them.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this exploratory analysis of multimodal vascular phenotypes within the OphtalmoLaus cohort, we examined how imaging-derived vascular features relate across retinal and peripheral vascular territories. The global correlation structure was dominated by stronger within-modality than between-modality associations, with a higher median absolute correlation within modalities than across modalities (median |r| = 0.12 vs 0.05). This indicates that vascular features are organised into modality-specific domains rather than forming a single, strongly coupled network across anatomical territories.\u003c/p\u003e \u003cp\u003eCross-territory correlations were generally modest and heterogeneous, with only a small proportion of feature pairs showing moderate associations. Retinal structural and perfusion-related features showed limited correspondence with oral and peripheral microcirculatory measures, suggesting that retinal imaging captures only part of the broader systemic vascular phenotype. These findings support a model of selective rather than global coupling across vascular beds.\u003c/p\u003e \u003cp\u003eWithin peripheral microcirculation, sublingual and sublabial features showed the clearest cross-bed concordance, with moderate associations observed for homologous features such as average capillary length (r\u0026thinsp;=\u0026thinsp;0.63). At the same time, inverse relationships between capillary length and segment-based descriptors highlight the non-redundant nature of these measures. Given the smaller sample size for oral microcirculatory imaging, these findings should be interpreted cautiously.\u003c/p\u003e \u003cp\u003eWithin the retina, OCTA and CFI showed the strongest cross-modality alignment, particularly for vascular geometry-related features, although overall correlations remained modest. The incomplete correspondence between OCT, OCTA, and CFI further supports the view that structural, perfusion, and large-vessel phenotypes represent distinct but related aspects of vascular organisation.\u003c/p\u003e \u003cp\u003eThis study has several limitations. It is cross-sectional and correlation-based, and does not account for systemic covariates such as age, sex, or cardiovascular risk factors. The effective sample size varied across modality pairs, and part of the feature set included exploratory outputs from an extended version of the CapX pipeline. These findings should therefore be interpreted as an initial characterisation of cross-territory vascular organisation rather than definitive evidence of biological coupling.\u003c/p\u003e \u003cp\u003eOverall, multimodal vascular phenotyping reveals a structured but only partially shared organisation of vascular traits across anatomical territories. Retinal and peripheral microcirculatory imaging provide overlapping yet non-redundant information, supporting the value of integrated multimodal approaches for studying vascular biology.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eMultimodal vascular phenotyping in OphtalmoLaus reveals a structured but only partially shared organisation of vascular traits across anatomical territories. Correlations were consistently stronger within-modality than between-modality correlations (median |r| = 0.12 vs 0.05), supporting the presence of modality-specific vascular phenotype domains. Cross-territory associations were generally modest, with limited correspondence between retinal and peripheral microcirculatory features. These findings suggest that vascular beds are selectively rather than globally coupled, and that retinal and peripheral imaging provide overlapping yet non-redundant information for the characterisation of microvascular organisation.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCFI, colour fundus imaging; CVI, choroidal vascularity index; FAZ, foveal avascular zone; GCL+IPL, ganglion cell layer + inner plexiform layer; INL+OPL, inner nuclear layer + outer plexiform layer; OCT, optical coherence tomography; OCTA, optical coherence tomography angiography; RNFL, retinal nerve fibre layer.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Swiss Personalized Health Network (grant 2018DRI13 to Thomas J. Wolfensberger), the Claire and Selma Kattenburg Foundation (Prix Kattenburg 2022 to Mattia Tomasoni, 2023 to Ciara Bergin and Mattia Tomasoni and 2025 to Mattia Tomasoni), and the Swiss National Science Foundation (grant no. CRSII5 209510 to Sven Bergmann).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eApproved by CER-VD (PB_2019-00168). Written informed consent was obtained from all participants.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eInce C, Boerma EC, Cecconi M, De Backer D, Shapiro NI, Duranteau J et al (2018) Second consensus on the assessment of sublingual microcirculation in critically ill patients: results from a task force of the European Society of Intensive Care Medicine. Intensive Care Med 44:281\u0026ndash;299\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan Y, Dong M, Wen S, Yuan X, Zhou L (2024) Retinal microcirculation: A window into systemic circulation and metabolic disease. Exp Eye Res 242:109885\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWong TY, Islam FMA, Klein R, Klein BEK, Cotch MF, Castro C et al (2006) Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (MESA). Invest Ophthalmol Vis Sci 47:2341\u0026ndash;2350\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGeechan K, Liew G, Macaskill P, Irwig L, Klein R, Klein BEK et al (2009) Meta-analysis: retinal vessel caliber and risk for coronary heart disease. Ann Intern Med 151:404\u0026ndash;413\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeidelmann SB, Claggett B, Bravo PE, Gupta A, Farhad H, Klein BE et al (2016) Retinal Vessel Calibers in Predicting Long-Term Cardiovascular Outcomes: The Atherosclerosis Risk in Communities Study. Circulation 134:1328\u0026ndash;1338\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGeechan K, Liew G, Macaskill P, Irwig L, Klein R, Klein BEK et al (2009) Prediction of incident stroke events based on retinal vessel caliber: a systematic review and individual-participant meta-analysis. Am J Epidemiol 170:1323\u0026ndash;1332\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOptical coherence tomography angiography (2024) of the retina and choroid in systemic diseases. Prog Retin Eye Res 103:101292\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRetinal Vascular Signs (2011) A Window to the Heart? Revista Espa\u0026ntilde;ola de Cardiolog\u0026iacute;a. (English Edition) 64:515\u0026ndash;521\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMousavi S, Garjani A, Elwakil A, Brock LP, Dherse A, Forestier E et al (2024) Cohort Builder: A Software Pipeline for Generating Patient Cohorts with Predetermined Baseline Characteristics from Medical Records and Raw Ophthalmic Imaging Data. Stud Health Technol Inform 316. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3233/SHTI240613\u003c/span\u003e\u003cspan address=\"10.3233/SHTI240613\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Zanet S, Ciller C, Wolf S, Sznitman R, Pathological (2017) OCT Retinal Layer Segmentation Using Branch Residual U-Shape Networks. Medical Image Computing and Computer Assisted Intervention\u0026thinsp;\u0026ndash;\u0026thinsp;MICCAI 2017; 294\u0026ndash;301\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElwakil A, Martin T, Quiros JV, Liefers B, Kropp M, Duriez A et al CapillaryX: A Fine-Tunable Pipeline for OCTA Segmentation and Feature Extraction. 2025 [cited 24 Mar 2026]. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21203/rs.3.rs-8176394/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-8176394/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVargas Quiros J, Liefers B, van Garderen KA, Vermeulen JP, Klaver C (2025) VascX Models: Deep Ensembles for Retinal Vascular Analysis From Color Fundus Images. Transl Vis Sci Technol 14:19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeloni I, Elwakil A, Aur\u0026eacute;lia G, Flavie T, Navarro A, Bagatella L et al Cohort profile: OphtalmoLaus, an extension of the CoLaus|PsyCoLaus cohort to investigate the relationships between ocular, cardiovascular, and cognitive parameters. 2025 [cited 24 Mar 2026]. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21203/rs.3.rs-7660406/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-7660406/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\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":"Microcirculation, OCT angiography, retinal biomarkers, multimodal imaging, vascular phenotyping, correlation analysis","lastPublishedDoi":"10.21203/rs.3.rs-9243114/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9243114/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e \u003cp\u003eWhether vascular phenotypes measured in one anatomical territory reflect those observed in other vascular beds remains unclear. Retinal imaging is increasingly used as a source of systemic vascular biomarkers, yet direct comparisons with oral and peripheral microcirculatory phenotypes within the same individuals remain limited.\u003c/p\u003e\u003ch2\u003eMethods:\u003c/h2\u003e \u003cp\u003eWe investigated the cross-territory organisation of multimodal vascular phenotypes in a deeply phenotyped subset of the OphtalmoLaus cohort. The analysis included 2,523 participants with ocular imaging available after quality control, of whom 82 also underwent sublingual and sublabial microcirculatory imaging. Vascular features were derived from structural optical coherence tomography (OCT), OCT angiography (OCTA), colour fundus imaging (CFI), sublingual imaging, sublabial imaging, and an exploratory family of capillary network-derived OCTA features. Pairwise Pearson correlation coefficients were computed across all available feature pairs to examine within-modality structure and cross-modality concordance.\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e \u003cp\u003eThe global correlation matrix showed a clear block-diagonal organisation, with stronger within-modality than between-modality correlations. Distinct internal clustering was observed across OCT, OCTA, CFI, and exploratory capillary network-derived feature families, indicating coherent modality-specific vascular phenotype domains. Cross-modality correlations were generally weaker and more heterogeneous, arguing against a single strongly coupled vascular phenotype shared across all territories. Among extra-ocular comparisons, sublingual and sublabial features showed the clearest cross-bed concordance, whereas retinal structural and perfusion-related phenotypes showed only limited correspondence with oral and peripheral microcirculatory measures. Within the retinal domain, OCTA and CFI demonstrated the strongest cross-modality alignment, consistent with partial overlap between perfusion- and fundus-derived vascular phenotypes.\u003c/p\u003e\u003ch2\u003eConclusions:\u003c/h2\u003e \u003cp\u003eMultimodal vascular phenotyping in OphtalmoLaus reveals a structured but only partially shared organisation of vascular traits across anatomical territories. These findings support a model of selective, territory-specific coupling across vascular beds and suggest that retinal and peripheral microcirculatory imaging provide overlapping yet non-redundant information.\u003c/p\u003e","manuscriptTitle":"Partial concordance of vascular phenotypes across retinal and peripheral microvascular beds in the OphtalmoLaus cohort","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-30 05:04:32","doi":"10.21203/rs.3.rs-9243114/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":"4709c083-7157-43d8-aeac-90c51264fe31","owner":[],"postedDate":"March 30th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65248815,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2026-03-30T05:04:32+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-30 05:04:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9243114","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9243114","identity":"rs-9243114","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}