PREDiCTOR Study to Assess the Effectiveness of AI in Producing Psychiatric Objective Measures

René S. Kahn, MD, PhD, is the Chair of Psychiatry at Icahn School of Medicine at Mount Sinai, and has spent his career working in psychiatry after training in both psychiatry and neurology. Kahn’s research focuses on schizophrenia, mainly working in brain imaging and clinical trials comparing various medications in patients with schizophrenia.

Mount Sinai announced in September 2024 that Kahn would be 1 of 3 investigators leading a new study, joining Cheryl Corcoran, MD, and Guillermo Cecchi, PhD. The Phenotypes Reimagined to Define Clinical Treatment and Outcome Research (PREDiCTOR) study aims to leverage advances in artificial intelligence (AI) to address the lack of objective measures in psychiatry.¹ A press release from Mount Sinai said the study will be conducted in collaboration with researchers from Harvard, Johns Hopkins, Columbia, and Carnegie Mellon universities, as well as Deliberate AI. Kahn sat down with Psychiatric Times to discuss the study and the potential AI has for future clinical use.

“The problem in psychiatry is that most of our assessments are subjective based on the conversation we have with the patient, based on the interpretation of that conversation, and based on our psychiatric interview, which again, is subjective,” Kahn said. Kahn said there is a feeling that AI can help develop objective measures that other fields of medicine have when diagnosing a patient. A recording device will be used to record the psychiatric interview, which will be fed to the AI program to predict the outcome in psychiatric patients who make the first visit to the psychiatric outpatient clinic.

A press release from Mount Sinai said the study will incorporate behavioral data from clinical interviews, at-home data captured on smartphones, and cognitive testing. "The team 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 press release said.¹

The study is being funded by a $20 million grant from the National Institute of Mental Health. The study has 3 clinically relevant, hard outcomes that investigators are interested in: is the patient coming back for treatment, are they being hospitalized, and are they going to the emergency room for any issue. Kahn said the study will be using outpatient clinics in the Mount Sinai system, with 4 hospitals, 8 clinics, 4 adult clinics, and 4 child and adolescent clinics. Eligible patients include those between the ages of 15 and 45 who come in for their first outpatient visit. The study will use 2 samples, with a training sample of 1500 subjects and a validation sample of 600 subjects, totaling 2100 participants.

Kahn said the biggest challenge investigators anticipated was patients' willingness to be recorded, but a pilot study with 31 participants found that 29 of the 31 included gave consent. “The other issue is that also the provider, the clinicians, need to provide consent because they will be part of the recording,” Kahn said. Investigators did not do a pilot study for this aspect, but Kahn said it was reassuring to see how willing patients were to be involved.

Dr Kahn is the Esther and Joseph Klingenstein Professor, System Chair of Psychiatry, and Inaugural Director of the Blau Center at the Icahn School of Medicine at Mount Sinai.

1. Mount Sinai Health System and IBM Research launch effort that leverages artificial intelligence and behavior data to improve mental health care for young people. Press release. Mount Sinai. September 12, 2024. Accessed February 27, 2025. https://www.mountsinai.org/about/newsroom/2024/mount-sinai-health-system-and-ibm-research-launch-effort-that-leverages-artificial-intelligence-and-behavioral-data-to-improve-mental-health-care-for-young-people