Background. Cardiovascular diseases (CVDs) remain the leading cause of premature death globally. Military personnel, including veterans, represent a high-risk group due to chronic stress and an increased burden of vascular risk factors. Retinal microvascular changes are known to reflect systemic vascular health and may serve as noninvasive biomarkers for early cardiovascular risk assessment. Objective. To evaluate the diagnostic performance of artificial intelligence (AI) -based fundus image analysis compared to traditional ophthalmoscopy in detecting retinal microvascular abnormalities associated with cardiovascular risk. Methods. The study included 391 participants, of whom 287 were active or recently discharged military personnel, and 104 were civilian controls. Retinal images were acquired using high-resolution, non-mydriatic fundus cameras and analyzed via the Retina-based Microvascular Health Assessment System (RMHAS), an AI-powered, validated platform. Features assessed included venular widening, arteriolar narrowing, arteriovenous ratio (AVR), vessel tortuosity, and caliber asymmetry. Diagnostic sensitivity, specificity, and area under the ROC curve (AUC) were calculated for various screening models. Results. AI demonstrated significantly higher detection rates of all key retinal signs compared to ophthalmoscopy (e.g., venular dilation: 43,0% vs. 17,0%). A combined model integrating AI analysis with blood pressure and BMI achieved the highest diagnostic performance (sensitivity: 83,3%; specificity: 80,2%; AUC: 0.84). AI alone outperformed traditional risk assessment based on clinical metrics. Conclusion. AI-based retinal analysis enables more sensitive, standardized, and early detection of microvascular changes indicative of cardiovascular risk.

Retinal imaging, Artificial intelligence, Cardiovascular risk, Military personnel

Wong DYL, Lam MC, Ran A, Cheung CY. Artificial intelligence in retinal imaging for cardiovascular disease prediction: current trends and future directions. Curr Opin Ophthalmol. 2022 Sep 1;33(5):440-446. doi: 10.1097/ICU.0000000000000886

Zhang W, Guo X, Jiang X, Liu J, Han X, Guo C. RETINAL MICROVASCULAR CHANGES AND RISK OF CORONARY HEART DISEASE: A Systematic Review and Meta-Analysis. Retina. 2024 Feb 1;44(2):333-344. doi: 10.1097/IAE.0000000000003959

Colcombe J, Mundae R, Kaiser A, Bijon J, Modi Y. Retinal Findings and Cardiovascular Risk: Prognostic Conditions, Novel Biomarkers, and Emerging Image Analysis Techniques. J Pers Med. 2023 Oct 31;13(11):1564. doi: 10.3390/jpm13111564

Arnould L, Meriaudeau F, Guenancia C, Germanese C, Delcourt C, Kawasaki R, Cheung CY, Creuzot-Garcher C, Grzybowski A. Using Artificial Intelligence to Analyse the Retinal Vascular Network: The Future of Cardiovascular Risk Assessment Based on Oculomics? A Narrative Review. Ophthalmol Ther. 2023 Apr;12(2):657-674. doi: 10.1007/s40123-022-00641-5

Syed MG, Trucco E, Mookiah MRK, Lang CC, McCrimmon RJ, Palmer CNA, Pearson ER, Doney ASF, Mordi IR. Deep-learning prediction of cardiovascular outcomes from routine retinal images in individuals with type 2 diabetes. Cardiovasc Diabetol. 2025 Jan 2;24(1):3. doi: 10.1186/s12933-024-02564-w

Iorga RE, Costin D, Munteanu-Dănulescu RS, Rezuș E, Moraru AD. Non-Invasive Retinal Vessel Analysis as a Predictor for Cardiovascular Disease. J Pers Med. 2024 May 9;14(5):501. doi: 10.3390/jpm14050501

Danielescu C, Dabija MG, Nedelcu AH, Lupu VV, Lupu A, Ioniuc I, Gîlcă-Blanariu GE, Donica VC, Anton ML, Musat O. Automated Retinal Vessel Analysis Based on Fundus Photographs as a Predictor for Non-Ophthalmic Diseases-Evolution and Perspectives. J Pers Med. 2023 Dec 29;14(1):45. doi: 10.3390/jpm14010045

Chikumba S, Hu Y, Luo J. Deep learning-based fundus image analysis for cardiovascular disease: a review. Ther Adv Chronic Dis. 2023 Nov 18;14:20406223231209895. doi: 10.1177/20406223231209895