AI Innovation in Mental Health Care

Medical Timeseries Networks (MTN), where Dr. Pettine serves as co-founder and CEO, develops AI-enabled systems that integrate real-time data from wearables, telemetry, and electronic health records to help clinical teams prioritize tasks and manage resources effectively.  These tools support workflow efficiency while keeping clinicians central to decision-making.  The University of Utah contributes through collaborative research and development, working alongside MTN through partnerships with the Spencer Fox Eccles School of Medicine Department of Biomedical Informatics, the Scientific Computing and Imaging Institute, Huntsman Mental Health Institute, and students in Biomedical Engineering.  Joint projects include clinical decision support models, electronic health record foundation models, biophysical digital twins, clinical note natural language processing, and synthetic label generation from EHR data.  Together, these collaborations help ensure that academic discovery and industry innovation inform and strengthen one another.

Dr. Pettine’s path to this work is deeply interdisciplinary.  While trained as a physician, he moved early into computational and systems neuroscience, spending more than a decade applying machine learning to complex biological and behavioral signals.  Today, he combines that expertise with health-systems research at the University of Utah, where he directs the Medical Machine Intelligence Lab.

Separate from MTN’s operational platform, Dr. Pettine leads the Medical Machine Intelligence Lab, which conducts foundational research at the intersection of neuroscience, data science, and medicine.  Investigators Dr. Brian Locke and Dr. Matthias Christenson mentor several students in the lab, contributing to the development of emerging researchers and supporting multidisciplinary innovation.

The lab’s research includes developing genomic foundation models to understand complex polygenic interactions in neuropsychiatric disease, designing sequence-based models that interpret clinical time series data for applications such as addiction, dementia, suicide prevention, and maternal health, harmonizing large and diverse medical datasets to enable reliable machine learning, and investigating the cognitive mechanisms underlying decision-making in clinical practice and neuropsychiatric conditions.  These efforts aim to advance scientific understanding and support more equitable, interpretable, and effective applications of AI in healthcare.

The Medical Machine Intelligence Lab recently released VERA, an open-source AI agent designed to programmatically replicate scientific papers.  VERA offers a promising approach to addressing the reproducibility crisis in scientific research by independently re-executing published methods and comparing outcomes.  While still in beta, the tool represents a significant step toward more transparent and verifiable scientific evaluation. Details are available in the project blog.

Across both academic inquiry and collaborative innovation, Dr. Pettine emphasizes that AI’s role should support and enhance human judgment.  By advancing operational tools through MTN and foundational research through the Medical Machine Intelligence Lab, his work reflects a commitment to responsible, human-centered innovation in mental health and healthcare delivery.

Learn more about the MTN work featured in TechBuzz News.