James Diao

About James Diao

James Diao grew up amid Chinese restaurants, street signs, and supermarkets in the largest Chinatown of Houston, Texas. He was first raised by his grandparents, who delighted in teaching him parodies of famous poems to recite. Later, when his grandparents returned to Jiangxian, James continued to spend the summers with them and his extended family. In the United States, James’s family later moved to Fort Bend, Texas, among the most diverse counties in the country, where he met classmates, teachers, and pastors with very different life experiences from his own.

Inspired by his experiences in the Texas Medical Center and his parents’ love of programming, James soon became interested in the intersection of health and technology. These interests drove James throughout his work at Apple, where he led projects to validate wearable health features in diverse populations; at PathAI, where he built deep learning models to predict molecular features from cancer histology slides; at the University of Cambridge, where he studied science policy as a Churchill Scholar; and at Harvard, where he researched the impacts of clinical guidelines on marginalized groups. James’s lead author research in NEJM and JAMA studied the use of race in clinical equations for kidney function, lung function, and cardiovascular disease. His other works on machine learning and precision medicine have appeared in Cell, Cell Systems, Human Molecular Genetics, and Nature Communications.

James is currently a cardiology fellow at Massachusetts General Hospital, postdoctoral research fellow at the Harvard Medical School Department of Biomedical Informatics, and associate editor at NEJM AI. He is a proud alumnus of the Harvard-MIT Program in Health Sciences and Technology, where he was the 22nd person at Harvard Medical School to graduate with highest honors (summa cum laude). Following his medical and scientific training, James hopes to continue exploring the capabilities and limits of AI for quantifying phenotypes, forecasting disease, and informing clinical decision-making.