Researchers from Washington University School of Medicine in St. Louis, the University of Washington in Seattle, and Genentech have developed OCTCube-M, an artificial intelligence system designed to accelerate the analysis of three-dimensional retinal scans and improve early diagnosis of vision-threatening conditions. The study, published in Nature Biomedical Engineering, addresses the clinical bottleneck of manually processing hundreds of images generated per optical coherence tomography scan. OCTCube-M comprises a suite of three AI models trained on over 26,000 three-dimensional optical coherence tomography datasets, encompassing 1.62 million individual retinal slices. By transitioning from traditional two-dimensional imaging to volumetric analysis, the system captures disease progression across the entire depth of retinal tissue. When evaluated against existing models, OCTCube-M demonstrated a four to six percent accuracy improvement in identifying eight major retinal diseases, including age-related macular degeneration. This performance gain translates to detecting an additional 43 to 60 cases per 1,000 screened patients across diverse clinical sites and demographic groups. The technology integrates optical coherence tomography with infrared retinal imaging and fundus autofluorescence to construct a comprehensive diagnostic view. This multimodal approach significantly enhances prognostic capabilities, particularly for geographic atrophy, a severe form of macular degeneration affecting approximately five million individuals worldwide. The AI model outperformed current state-of-the-art prognostic tools by nearly 50 percent in predicting disease progression rates. Faster and more accurate progression tracking enables more efficient clinical trial design, potentially reducing development costs and accelerating therapeutic approvals. Beyond ophthalmology, the system demonstrates remarkable systemic health inference capabilities. Analyzing retinal vasculature, which shares anatomical and developmental characteristics with the kidneys and cerebral circulatory systems, the AI successfully predicted risks for heart attack, stroke, and kidney failure. Lead researcher Aaron Lee, MD, emphasizes that transforming routine eye exams into comprehensive health screenings could facilitate earlier intervention for patients otherwise diagnosed at advanced disease stages. Future development will focus on expanding training datasets to encompass broader patient populations, additional pathologies, and advanced imaging modalities. The continued refinement of OCTCube-M positions AI-driven retinal analysis as a scalable solution for reducing diagnostic latency, optimizing treatment personalization, and enhancing global eye care infrastructure.