We build Artificial Intelligence for Medical Applications

The Rajpurkar Lab has started up in the Department of Biomedical Informatics at Harvard Medical School.

Few-Shot Medical AI

We are dramatically expanding the scope of medical AI applications with novel self-supervision, multimodal, and transfer learning approaches.

Real-World Medical Data Curation

We are partnering with hospitals worldwide to transform the basis for AI development from simplistic, homogenous datasets to real-world, heterogeneous datasets.

Medical AI Deployment

We are collaborating with clinicians to implement medical AI in clinical workflows, improving downstream impact on decision making and clinical outcomes.

Welcome

Greetings! We are a new lab at Harvard DBMI led by 🔗 Dr. Pranav Rajpurkar. We work on building reliable AI technologies for medical decision making. We have developed deep learning algorithms that can read medical images at the level of experts, we have built large-scale open medical datasets, and we have demonstrated the positive and negative effects of AI on medical decision making.

What will it take to make medical AI part of routine clinical care? We believe that algorithm development for limited labeled data settings, real-world dataset curation at scale, and the design of effective clinical deployment setups are critical directions. I welcome you to join us in our mission.

Students

We welcome undergraduate, graduate, and post-doctoral students and research volunteers with a background in any one of artificial intelligence, software engineering, or medicine. By working with our lab, you will build solutions to important problems in medicine. You will learn and sharpen cutting-edge AI through your research, your peers, and reading groups. You will work together in small teams with other scientists, engineers, and clinicians, and receive mentorship to author a research paper.

We expect that students will be able to commit at least 20 hours per week to research and be able to commit at least 6 months of research with the lab. If you are interested in applying, please email your CV and brief description of interest to rajpurkarlab-apply at hms.harvard.edu, specifying whether you are Harvard undergrad or graduate student, volunteer researcher, visiting scholar, or postdoc in the subject line.

Ekin Tiu

Ellie Talius

Pujan Patel

Join Us

Collaborators

We cultivate collaborations with academic institutions, hospitals and industry. We are currently interested in collaborating to deploy medical imaging AI models. We are also partnering with hospitals and industry to create large multi-institutional imaging datasets. If you are interested in a possible collaboration with the Rajpurkar lab, please email rajpurkarlab-apply at hms.harvard.edu with the subject line Possible Research Collaboration.