OCULOMICS: AN ARTIFICIAL INTELLIGENCE–DRIVEN FRAMEWORK FOR RETINAL IMAGING IN SYSTEMIC DISEASE ASSESSMENT

The retina offers a unique, non-invasive window into systemic health owing to its shared embryological origin with the central nervous system and its capacity for direct visualisation of microvascular and neural structures. Recent advances in retinal imaging modalities, including colour fundus photography, optical coherence tomography, optical coherence tomography angiography, fundus autofluorescence, and emerging molecular techniques, have expanded the utility of ocular biomarkers beyond traditional ophthalmic disease assessment. The convergence of these technologies with artificial intelligence has led to the emergence of oculomics, a field focused on extracting systemic health information from retinal data. Artificial intelligence-enabled analysis supports automated, scalable, and reproducible extraction of complex retinal features and facilitates multimodal data integration, enhancing detection of subtle structural, vascular, and metabolic alterations. Retinal biomarkers derived through oculomics have shown associations with a wide range of systemic conditions, including cardiovascular disease, diabetes mellitus, neurodegenerative disorders, renal disease, and inflammatory and autoimmune conditions, underscoring the potential of the retina as a surrogate marker of systemic pathology. Despite its promise, challenges related to model interpretability, algorithmic bias, data privacy, and external validation must be addressed to enable responsible clinical translation. This review summarises current advances in retinal imaging based oculomics, highlighting the role of artificial intelligence–driven multimodal integration in systemic disease assessment and its potential contribution to precision medicine and preventive healthcare.