About Our Research
The highly collaborative Medical Machine Intelligence Lab, led by Warren Pettine, MD, investigates a broad range of questions in neuropsychiatric, mental health, and medicine centered on the learning of patterns from information in the world.
Research Areas
Genomic Foundation Models and Neuropsychiatric Disease
The large language models powering ChatGPT have the potential to unlock similar advances in treating genetic conditions. We investigate the use of innovating genomic foundation models to uncover complicated polygenetic interactions. In particular, we focus on predicting the progression of Alzheimer’s dementia.
Medical Sequence Models
Health status changes over time, requiring models that detect complicated, noisy patterns. We develop novel architectures for detecting patterns in sequential medical data, such as telemetry or electronic health records. Applications include addiction, dementia, suicide prevention and maternal health.
Medical Data Harmonization
While is an ever-growing abundance of rich medical data, different formats and properties hinder its ability to advance medicine. We undertake large efforts to harmonize data across sources, optimizing data for use in machine learning.
Decision-making
Machine are incredible tools to aid in human decision making, and to gain insight into underlaying cognitive mechanisms. We use computational models to investigate the properties of decision-making, how it can be optimized in medical practice, and how it can be altered in neuropsychiatric conditions.
Academia-Industry Collaborations
We actively collaborate with industry partners to uncover critical, impactful insights from data. If you have an interesting data problem and seek collaboration with an academic medical center, please reach out.
Affiliations
Meet our research team
Principal Investigator
Warren Woodrich Pettine, MD
Dr. Pettine is an Assistant Professor of Psychiatry at Huntsman Mental Health Institute. He is adjunct faculty in the Department of Biomedical Engineering and has affiliations with the Scientific Computing and Imaging Institute and the Interdisciplinary Neuroscience Program. He received computational neuroscience training through research positions at Stanford, NYU, and Yale.
Investigators
Matthias Christenson, PhD
Dr. Christenson is adjunct at the University of Utah and Principal Machine Learning Scientist at MTN. He trained in computational neuroscience at University College London and Columbia University and has extensive industry experience in the application of machine intelligence to biomedicine.
Charles Kemmler, MD, PhD
Dr. Charles Kemmler is an Assistant Professor at the University of Pennsylvania. He has a Ph.D. and M.D. from the University of Colorado. He is an expert in the application of machine intelligence to clinical practice.
Pranav Koirala, MD
Dr. Koirala is an Adjunct Assistant Professor at the University of Maryland. He practiced as an expedition medicine physician in the Himalayas, and has extensive industry experience in medical machine learning.
Brian W. Locke, MD MSCI
Dr. Locke is an Assistant Professor of Pulmonary and Critical Care at Intermountain Health. He had data science training from the University of Colorado and the University of Utah. Dr. Locke is an expert in the application of machine intelligence to clinical decisions.
Collaborators
Jorie Butler, PhD
Hilary Coon, PhD
Anna Docherty, PhD
Shireen Elhabian, PhD
Vincent Kopplemans, PhD
Paul Rosen, PhD
Andrey Shabalin, PhD
Disclosure: The senior research team of the Medical Machine Intelligence Lab hold a financial interest in Mountain Biometrics, Inc. (MTN).
Selected Publications
Contact
Warren Pettine, MD
Principal Investigator