Pranav Rajpurkar, Associate Professor of Biomedical Informatics, Harvard Medical School

Pranav Rajpurkar is an associate professor of biomedical informatics at Harvard Medical School. He wanted to be a scientist at 13, walked into Andrew Ng's lab as a Stanford freshman, and as a doctoral student co-led CheXNet, a model that read chest X-rays for pneumonia. He now leads the Rajpurkar Lab and co-founded a2z Radiology AI.

Joining Andrew Ng's Lab as a Freshman

His parents gave him a love of science and reading early. "During freshman orientation at Stanford, I heard Andrew Ng from Stanford's artificial intelligence lab give a talk on AI," he recalled. He emailed Ng, joined the lab that year, and spent his undergraduate years on AI for autonomous driving. He went on to earn a bachelor's, a master's, and a PhD in computer science at Stanford.

Turning to Medicine

In the second year of his PhD he moved to medicine, starting with models that detect abnormal heart rhythms in electrocardiograms. As a graduate student in Ng's group he co-led CheXNet with Jeremy Irvin, trained on 112,120 chest X-rays released by the NIH Clinical Center. On 420 images annotated by four Stanford radiologists, CheXNet agreed with their majority vote more often than the individual radiologists did. CheXpert, from the same line of work, became a large labeled set other labs train on, and he also worked on SQuAD, a reading benchmark of more than 100,000 questions.

Leading a Lab at Harvard

His Harvard group works on multimodal medical AI, clinical agents, and studies of when pairing a clinician with a model helps or fails. More than 90,000 people have enrolled in his AI for Medicine course.

He co-founded a2z Radiology AI. Its first product, a2z-Unified-Triage, has FDA clearance to prioritize adult abdomen and pelvis CT studies when a suspected finding falls in one of seven conditions.

Links

← All people · Featured in Voices of Healthcare: 30 People Changing Health Tech (2026)