{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T15:56:23Z","timestamp":1787932583108,"version":"build-2784847793"},"reference-count":26,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T00:00:00Z","timestamp":1786492800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>We developed and validated RetinalVNG-Net, a multimodal deep learning framework for simultaneous risk stratification of hypertensive retinopathy, diabetic retinopathy, and neurodegenerative-associated retinal changes from integrated fundus photography, optical coherence tomography, and clinical metadata.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      This retrospective multi-center study included 2,740 subjects from three independent ophthalmology centers (January 2019\u2013December 2023). Centers A and B (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f2,220) constituted the development set, with a stratified 15% subset (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f333) reserved for hyperparameter tuning and the remainder (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f1,887) used for five-fold cross-validation. Center C (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f520) served as a geographically distinct, device-heterogeneous external test set. RetinalVNG-Net employs a RETFound ViT-Large fundus encoder, a dual-stream Optical Coherence Tomography (OCT) branch (ResNet-3D-18 for volumetric B-scans and 2D-CNN for layer thickness maps), and a tabular transformer for metadata, fused via cross-modal attention. An auxiliary regression head outputs a continuous Retinal Biological Age Gap (RBAG) score as an interpretable severity biomarker.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Internal cross-validation yielded macro-averaged AUC-ROC 0.957 (\u00b10.008), sensitivity 0.913, specificity 0.941, and F1 0.908 across four classes. On the external test set, macro-averaged AUC-ROC reached 0.944 (95% CI 0.922\u20130.958), with per-class AUCs of 0.963 (hypertensive retinopathy), 0.951 (diabetic retinopathy), 0.924 (neurodegenerative changes), and 0.938 (controls). RetinalVNG-Net significantly outperformed the best single-modality fundus model (macro-AUC 0.944 vs. 0.901;\n                      <jats:italic>p<\/jats:italic>\n                      \u202f&amp;lt;\u202f0.001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>RetinalVNG-Net demonstrates promising robustness and generalizability for simultaneous multimodal risk stratification of vascular and neurodegenerative retinal changes across diverse devices and settings, supporting further evaluation as a tool for risk stratification of retinal manifestations associated with systemic vascular and neurodegenerative disease. Prospective longitudinal studies would be required to establish value for early or predictive detection.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fneur.2026.1885303","type":"journal-article","created":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T04:37:58Z","timestamp":1786509478000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Artificial intelligence-integrated multimodal retinal imaging for early detection and risk stratification of systemic vascular and neurodegenerative diseases"],"prefix":"10.3389","volume":"17","author":[{"given":"Kaili","family":"Du","sequence":"first","affiliation":[{"name":"Eye Genebank, He University","place":["Shenyang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuqiang","family":"Zheng","sequence":"additional","affiliation":[{"name":"Eye Genebank, He University","place":["Shenyang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhuoshi","family":"Wang","sequence":"additional","affiliation":[{"name":"Eye Genebank, He University","place":["Shenyang, China"]},{"name":"He Eye Specialists Hospital","place":["Shenyang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,12]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"344","DOI":"10.1016\/S1474-4422(24)00038-3","article-title":"Global, regional, and national burden of disorders affecting the nervous system, 1990\u20132021: a systematic analysis for the global burden of disease study 2021","volume":"23","author":"Steinmetz","year":"2021","journal-title":"Lancet Neurol"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1186\/s40662-024-00384-3","article-title":"Prognostic potentials of AI in ophthalmology: systemic disease forecasting via retinal imaging","volume":"11","author":"Tan","year":"2024","journal-title":"Eye Vision"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"3223","DOI":"10.7150\/thno.100786","article-title":"Artificial intelligence-enhanced retinal imaging as a biomarker for systemic diseases","volume":"15","author":"Wang","year":"2025","journal-title":"Theranostics"},{"key":"ref4","doi-asserted-by":"crossref","first-page":"103089","DOI":"10.1016\/j.eclinm.2025.103089","article-title":"Use of artificial intelligence with retinal imaging in screening for diabetes-associated complications: systematic review","volume":"81","author":"Yang","year":"2025","journal-title":"EClinicalMedicine"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"e14601","DOI":"10.1002\/alz.14601","article-title":"A deep-learning retinal aging biomarker for cognitive decline and incident dementia","volume":"21","author":"Sim","year":"2025","journal-title":"Alzheimers Dement"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1097\/ICU.0000000000001174","article-title":"Artificial intelligence oculomics for systemic health and longevity medicine: 2025 and beyond","volume":"36","author":"Yao","year":"2025","journal-title":"Curr Opin Ophthalmol"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1167\/tvst.13.2.16","article-title":"Deep learning and machine learning algorithms for retinal image analysis in neurodegenerative disease: systematic review of datasets and models","volume":"13","author":"Bahr","year":"2024","journal-title":"Transl Vis Sci Technol"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"104788","DOI":"10.1016\/j.pdpdt.2025.104788","article-title":"Retinal imaging and artificial intelligence: a systematic review and meta-analysis of diagnostic techniques for neurodegenerative diseases","volume":"55","author":"Ganji","year":"2025","journal-title":"Photodiagn Photodyn Ther"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1317","DOI":"10.1038\/s41467-024-55635-7","article-title":"Multi-omic spatial effects on high-resolution AI-derived retinal thickness","volume":"16","author":"Jackson","year":"2025","journal-title":"Nat Commun"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1036","DOI":"10.1097\/MS9.0000000000004391","article-title":"AI-powered oculomics for early diagnosis of neurodegenerative diseases through retinal microvasculature analysis","volume":"88","author":"Rahman","year":"2026","journal-title":"Ann Med Surg"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"5083","DOI":"10.1038\/s41598-026-35383-y","article-title":"Non-invasive screening of Alzheimer\u2019s disease via label-free tri-spectral retinal imaging","volume":"16","author":"Salajkov\u00e1","year":"2026","journal-title":"Sci Rep"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"3001","DOI":"10.1038\/s41598-025-32776-3","article-title":"Machine learning assessment of retinal blood flow links metabolic dysfunction and accelerated microvascular aging","volume":"16","author":"Magi","year":"2026","journal-title":"Sci Rep"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"2150","DOI":"10.3390\/biomedicines12092150","article-title":"Retinal imaging-based oculomics: artificial intelligence as a tool in the diagnosis of cardiovascular and metabolic diseases","volume":"12","author":"Ghenciu","year":"2024","journal-title":"Biomedicine"},{"key":"ref14","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1038\/s41746-025-01588-0","article-title":"Vision transformer based interpretable metabolic syndrome classification using retinal images","volume":"8","author":"Lee","year":"2025","journal-title":"npj Digital Med"},{"key":"ref15","doi-asserted-by":"crossref","first-page":"367","DOI":"10.1038\/s41746-025-01699-8","article-title":"Artificial intelligence-derived retinal age gap as a marker for reproductive aging in women","volume":"8","author":"Miao","year":"2025","journal-title":"npj Digital Med"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1038\/s41586-023-06555-x","article-title":"A foundation model for generalizable disease detection from retinal images","volume":"622","author":"Zhou","year":"2023","journal-title":"Nature"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"6862","DOI":"10.1038\/s41467-025-62123-z","article-title":"Training a high-performance retinal foundation model with half-the-data and 400 times less compute","volume":"16","author":"Engelmann","year":"2025","journal-title":"Nat Commun"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"3637","DOI":"10.1038\/s41598-024-54251-1","article-title":"Deep learning predicts prevalent and incident Parkinson\u2019s disease from UK biobank fundus imaging","volume":"14","author":"Tran","year":"2024","journal-title":"Sci Rep"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1325","DOI":"10.1038\/s41551-024-01291-7","article-title":"Seeing ageing through retinal images","volume":"8","author":"P\u00e0mies","year":"2024","journal-title":"Nat Biomed Eng"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"11039","DOI":"10.1038\/s41467-025-66979-z","article-title":"Artificial intelligence-driven metabolomics of retinal nerve fibre layer to profile risks of mortality and cardiometabolic diseases","volume":"16","author":"Yang","year":"2025","journal-title":"Nat Commun"},{"key":"ref21","doi-asserted-by":"crossref","first-page":"4012","DOI":"10.1038\/s41598-025-34153-6","article-title":"Multimodal deep learning approaches for improving polygenic risk scores with imaging data","volume":"16","author":"Kim","year":"2026","journal-title":"Sci Rep"},{"key":"ref22","doi-asserted-by":"crossref","DOI":"10.2139\/ssrn.5331997","article-title":"Harnessing AI and retinal imaging to transform early diagnosis of systemic diseases: a new clinical paradigm","volume-title":"SSRN","author":"Raposo","year":"2025"},{"key":"ref23","doi-asserted-by":"crossref","first-page":"afac062","DOI":"10.1093\/ageing\/afac062","article-title":"Retinal age gap as a predictive biomarker of future risk of Parkinson\u2019s disease","volume":"51","author":"Hu","year":"2022","journal-title":"Age Ageing"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"100593","DOI":"10.1016\/S2666-7568(24)00089-8","article-title":"Application of a deep-learning marker for morbidity and mortality prediction derived from retinal photographs: a cohort development and validation study","volume":"5","author":"Nusinovici","year":"2024","journal-title":"Lancet Healthy Longev"},{"key":"ref25","doi-asserted-by":"crossref","first-page":"344","DOI":"10.1038\/s41746-025-01751-7","article-title":"A cross population study of retinal aging biomarkers with longitudinal pre-training and label distribution learning","volume":"8","author":"Yu","year":"2025","journal-title":"npj Digital Med"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1038\/s41746-024-01109-5","article-title":"RETFound-enhanced community-based fundus disease screening: real-world evidence and decision curve analysis","volume":"7","author":"Zhang","year":"2024","journal-title":"NPJ Digital Med"}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1885303\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T04:38:11Z","timestamp":1786509491000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1885303\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,12]]},"references-count":26,"alternative-id":["10.3389\/fneur.2026.1885303"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2026.1885303","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,12]]},"article-number":"1885303"}}