Lab profile
About Rajpurkar Lab
The lab develops medical AI systems that reason from clinical evidence across scans and records, rigorously tests them in simulated clinical environments before deployment, and teaches robots to physically assist clinicians in operating rooms and at bedsides. Their goal is to build AI that predicts patient futures from comprehensive medical data, simulate entire hospitals for AI testing, and create portable surgical assistants.
01 · Research focus
Research areas and highlights
Topics and recent work as stated on the lab’s own site.
- MedVersaA generalist model that reads and writes both images and text across seven kinds of medical imaging, with radiologists rating its reports as good as human-written ones in 64% of cases.
- CRAFT-MDEvaluates clinical language models through interactive patient encounters, finding GPT-4's diagnostic accuracy fell from 82% on written cases to 63% when taking history itself.
- Clinical Environment SimulatorProposes a persistent digital hospital where agent decisions alter patients, resources, and subsequent choices for comprehensive AI testing.
- Robot suture followingA low-cost robot arm trained on under an hour of demonstrations followed 23 of 25 stitches alongside a surgeon in open surgery.
- ReXGroundingCTConnects free-text radiology findings to expert-verified, pixel-level evidence in three-dimensional chest CT.
- Heart risk from head CTReads routine emergency head CTs to predict heart attack, stroke, and heart failure, outperforming standard PREVENT risk score.
- ReXSonoVQAThe first benchmark for video-based procedural ultrasound understanding, testing whether models understand probe movement, image quality, anatomy, and procedural intent.
02 · People
Leadership and researchers
People listed on the lab’s team pages, by role.
03 · Research intelligence
Research Intelligence evidence
Publications, grants and clinical trials linked to the lab’s researchers in Health AI Central’s research data.
Publications & preprints
- Repurposing adult AI radiograph models for neonatal care through continual learning: an international multicentre study.Nature biomedical engineering · 2026
- An explainable biomedical foundation model via large-scale concept-enhanced vision-language pretraining.Nature biomedical engineering · 2026
- Large reasoning models as thinking machines for medicine.Nature biomedical engineering · 2026
Active NIH grants
- Integrating Clinical Knowledge: A Multimodal AI System for Reliable Head CT Understanding and InterpretationPranav Rajpurkar · NIBIB · FY 2026
04 · Updates
Updates and lab news
News and events from the lab’s own site.
- Adult AI models adapted for newborns in Nature Biomedical EngineeringPublicationOct 13, 2026 · nature.com
- Three papers in the Machine Learning for Healthcare 2026 proceedingsPublicationSep 8, 2026 · proceedings.mlr.press
- SQuAD receives the 2026 ACL Test-of-Time Paper AwardAwardJul 14, 2026 · aclweb.org
- ReXGrounding Challenge at MICCAI 2026EventMay 1, 2026 · rexrank.ai
- Head CT features improve cardiovascular risk prediction in JACCPublicationApr 8, 2026 · jacc.org
- Generalist biological AI in Nature BiotechnologyPublicationMar 20, 2026 · nature.com
- Clinical Environment Simulator in Nature MedicinePublicationMar 12, 2026 · nature.com
- MedVersa generalist medical foundation model in NEJM AIPublicationMar 5, 2026 · ai.nejm.org
05 · Official presence
Lab sources and channels
The lab’s own pages and accounts.
06 · Related
Related labs and centers
Labs at the same institution and labs working on similar topics.