Visionary AI: Pioneering Diagnostic Tools to Improve Early Detection of Preeclampsia Worldwide

Liat Shenhav of New York University Grossman School of Medicine in the U.S. will develop a diagnostic platform, based on non-invasive retinal imaging, for prediction and detection of preeclampsia early in pregnancy. Previous results with a U.S.-based cohort of 1,400 pregnant women showed promise in using retinal vasculature features in the first trimester to predict the risk of preeclampsia. With local collaborators, the current study will recruit 2,000 pregnant women in clinical research centers in Belagavi and Nagpur in India, part of the Global Network for Women’s and Children’s Health Research. Each participant will get a retinal scan in the first and second trimester and be followed through pregnancy to determine clinical outcomes. An AI-based model will be developed to predict preeclampsia risk from the retinal scans, guided by parameters used in modeling for the U.S.-based cohort to help ensure generalizability.