Machine Learning for Health (ML4H)

Machine Learning for Health (ML4H) is the Broad Institute's clinical AI group. We build and translate machine learning methods that turn real-world clinical data, including electronic health records, medical imaging, wearables, clinical notes, and genomics, into earlier diagnosis, sharper risk prediction, and new insight into disease biology.

ML4H is a multidisciplinary team of ML researchers, engineers, and clinicians working closely with Mass General Hospital, Brigham and Women's Hospital, MIT, and other partners across the Broad.

What we do:

Advance core AI methods for medicine
New representation learning, multimodal, generative, and LLM methods built for the realities of biomedical data, including electronic health records, medical imaging, wearable and sensor data, clinical notes, and genomics.

Deliver models for diagnosis and risk prediction
Externally validated models for disease detection, subtyping, and outcome prediction, with a focus on rigorous evaluation and clinical relevance.

Connect clinical signals to biology
AI-derived phenotypes that uncover genetic and molecular drivers of disease and surface candidate therapeutic targets.

Build with clinicians and the community
We partner with clinicians, researchers, and engineers to ensure our work is trustworthy, interpretable, and designed for use in real-world settings.

Learn about our key projects and browse our publications to explore methods, applications, and results.

Collaborate with us

To discuss collaboration opportunities, contact ML4H’s director, Mahnaz Maddah. For general inquiries, email ml4h@broadinstitute.org.