Harvard Catalyst Profiles

Predicting Cardiac Risk Factors from Retinal Imaging The ophthalmic world was stunned when Google described detection of heretofore unknown signals of cardiac risk factors (gender, age, systolic BP, HbA1c, and smoking status) in retinal fundus images1. Other imaging modalities also have been found to contain unexpected signals of human health and even potential biases2. Our proposed project will extend this work by using retinal and optic nerve optical coherence tomography (OCT and OCT-A) images to look for signals of disease and risk factors for disease. OCTs contain high-resolution, 3-dimentional cross-sectional images which may be a rich source of information about systemic health including CV diseases. Medical students, residents, and fellows are welcome to join the project to gain knowledge about Ophthalmology, Cardiology, Data Science, Bioinformatics, Medical Image Analysis, Deep Learning /AI. Please contact michael_morley@MEEI. harvard.edu 1 Poplin, R., Varadarajan, V., Blumer, K., et al. Prediction of Cardiovascular Risk Factors from Retinal Fundus Photographs Using Deep Learning Nature Biomedical Engineering 2, 158-164, (2018) 2 https://www.thelancet.com/journals/landig/article/PIIS2589-7500(22)00063-2/fulltext