Research Seeks to Leverage AI, Behavioral Data to Improve Mental Health

Sept. 12, 2024
Mount Sinai-led study seeks to predict outcomes around mental health evaluation and treatment

Mount Sinai Health System and IBM Research have launched a research effort that aims to address the lack of objective measures in psychiatry by leveraging advances in artificial intelligence and incorporating behavioral data from clinical interviews, at-home data captured on smartphones, and cognitive testing. 

One goal of the Phenotypes Reimagined to Define Clinical Treatment and Outcome Research (PREDiCTOR) study is to predict outcomes such as treatment discontinuation, hospitalizations, and emergency room visits for young people who are seeking mental health evaluation and treatment. 

The study will be conducted in collaboration with researchers from Harvard, Johns Hopkins, Columbia, and Carnegie Mellon universities and Deliberate AI.

Funded by a $20 million grant from the National Institute of Mental Health (NIMH), the PREDiCTOR study team said it will use objective, scalable, and cost-effective measurements to define novel clinical signatures that can be used for individual-level prediction and clinical decision-making in treating mental health disorders.

The research project will be co-led by Rene Kahn, M.D., Ph.D., Chair of Psychiatry at the Icahn School of Medicine at Mount Sinai and Mount Sinai Health System; Cheryl Corcoran, M.D., Program Leader in Psychosis Risk for Icahn Mount Sinai; and Guillermo Cecchi, Ph.D., Director of the Computational Psychiatry and Neuroimaging groups in IBM Research. Collaborators from Harvard, Johns Hopkins, Columbia, and Carnegie Mellon universities and Deliberate.ai will contribute to the work as well.

“Every clinical visit provides a wealth of untapped behavioral data that includes spoken language, eye contact, and facial expressions from both the patient and clinician,” said Corcoran, Associate Professor of Psychiatry at Icahn Mount Sinai, in a statement. “With advancements in computational approaches, these behaviors can be operationalized and quantified through analysis of audiovisual data obtained from the recording of clinical interviews. Coupled with valid behavioral data derived from smartphones that track physical activity metrics like step count and distance traveled, geolocation, social interactions like text messages and phone calls, sleep patterns, and audio data from diaries, we can develop clinical signatures that are indicative of key outcomes.”

The team will develop novel clinical signatures in patients aged 15 to 30 years who seek treatment for the first time at one of six outpatient mental health clinics within the Mount Sinai Health System. These clinics serve an ethnically diverse community from various socioeconomic backgrounds and this particular age range represents a developmental window during which many disturbances of thought, emotion, and behavior emerge and when diagnoses and prognoses are often still unclear.

“Individualized prognosis and clinical decision-making during this critical period may have a profound impact on the lifetime trajectory of these young patients,” said Kahn, the Esther and Joseph Klingenstein Professor of Psychiatry at Icahn Mount Sinai, in a statement. “Our team possesses longstanding expertise in electronic health record analysis, audiovisual behavior collection, and computational analysis of these data. We also have extensive involvement in community engagement and a strong track record of collaboration, making us uniquely poised to carry out this important work.”

The study team will invite all new patients, irrespective of their initial diagnosis, to have audio and visual recordings made of their clinical visits over one year. Cognitive function, as one of the most robust and well-established general predictors of outcome, will be assessed at baseline, and patients will be assessed over the one-year period. By analyzing these rich behavioral datasets from routine clinical visits and smartphones, the team will develop clinical signatures for particularly clinically relevant events (outcomes) in young help-seeking people, namely treatment disengagement, emergency room visits and hospitalizations.

“Our goal is to gain a better understanding of what predicts whether young people stay in mental health treatment or drop out, and what predicts whether their symptoms worsen such that they need acute care in an emergency crisis center or hospital,” explained Cecchi in a statement. “We have shown in our research that artificial intelligence can be used to predict some outcomes in controlled experimental settings, but we believe that current advancements are powerful enough to be applied in the context of usual clinical practice.”   

This research effort is supported by the NIMH’s Individually Measured Phenotypes to Advance Computational Translation in Mental Health program, a new initiative focused on using behavioral measures and computational methods to define novel clinical signatures that can be used for individual-level prediction and clinical decision making in treating mental disorders. 

 

Sponsored Recommendations

How AI-Native Locating Intelligence Revolutionizes the RTLS market

Discover how leveraging an RTLS solution with artificial intelligence as the location engine can increase efficiency, improve safety, and elevate care without the compromises ...

Harnessing the True Power of Cultural, Clinical and Operational Data

Optimize healthcare performance by combining clinical, operational, and cultural insights. A deeper understanding of team factors improves care and resource management.

How Digital Co-Pilots for patients help navigate care journeys to lower costs, increase profits, and improve patient outcomes

Discover how digital care journey platforms act as 'co-pilots' for patients, improving outcomes and reducing costs, while boosting profitability and patient satisfaction in this...

5 Strategies to Enhance Population Health with the ACG System

Explore five key ACG System features designed to amplify your population health program. Learn how to apply insights for targeted, effective care, improve overall health outcomes...