From retinal imaging to population prevention

Oculomics-driven deep learning

Operating within the oculomics framework — the study of systemic disease through ocular imaging — we are developing a suite of deep learning models trained to detect retinal biomarkers associated with systemic cardiometabolic and cerebrovascular risk.

Our models extract structural and vascular retinal signatures from standard fundus photographs, correlating these signatures with incident stroke, hypertension, and cardiovascular events.

From lab to national prevention policy

Our work is designed to produce real-world clinical and policy outcomes, not research abstracts alone. Validated model outputs will directly inform screening protocols and prevention strategy at a national and European level.


Primary deployment target

Romania — National Prevention Programme EU Cardiovascular Strategy Point-of-care screening

Validated outputs will directly inform the Romanian National Health Prevention Programme for cerebro-cardio-vascular diseases and contribute evidence relevant to the European Union's cardiovascular prevention strategy. Deployment is designed for point-of-care settings — optometry practices and primary care — requiring no specialist ophthalmic infrastructure.

Because our system operates from standard fundus photographs captured in any community optometry or primary care setting, the pathway from positive screen to clinical follow-up is short and does not depend on specialist ophthalmic infrastructure. This positions retinal AI as a credible instrument for large-scale, equitable cardiovascular prevention — reachable by the populations who need it most.