Artificial intelligence-driven metabolomics of the retinal nerve fiber layer to profile risks of mortality and cardiometabolic diseases.

Cardiometabolic diseases remain the leading causes of global morbidity and mortality, with early detection often hindered by nonspecific symptoms and reliance on systemic biomarkers. The retinal nerve fiber layer (RNFL), a microvascular and neurodegenerative biomarker, is highly sensitive to systemic metabolic and vascular insults. Artificial intelligence (AI)-driven metabolomics integrates high-resolution RNFL imaging with circulating metabolite profiling, enabling precise risk stratification for mortality and cardiometabolic disease. By combining optical coherence tomography data with machine learning algorithms, this approach deciphers complex biochemical signatures and correlates them with systemic outcomes. Recent studies demonstrate that AI-powered RNFL metabolomics achieves superior sensitivity and specificity compared to conventional diagnostic tools, with applications extending to diabetes, hypertension, neurodegenerative disorders, and chronic kidney disease. However, challenges such as dataset bias, limited accessibility of metabolomic assays, and regulatory hurdles remain. Synthesizing current evidence, AI-driven RNFL metabolomics represents a transformative innovation in precision medicine, offering a scalable, noninvasive pathway for early detection, personalized care, and improved survival outcomes in cardiometabolic disease.