October 6, 2025—BRONX, NY—Serious mental illnesses (SMI) take a tremendous toll on individuals, their friends and family, and society as a whole. SMI such as schizophrenia, bipolar disorder, and major depressive disorder contribute to poverty, unemployment, and homelessness, and can lead to hospitalization and suicide. Predicting when intensive intervention is needed in individual cases is a major unmet mental healthcare need.
Albert Einstein College of Medicine has received an $18 million grant from the National Institutes of Health (NIH) to use AI, cognitive monitoring, and psychiatric symptoms to determine when someone diagnosed with an SMI needs more intensive support. The project will develop prediction algorithms using AI and a novel cognitive assessment tool to help identify those at high risk for a crisis so they can be offered interventions to prevent symptom escalation, improve recovery time, and reduce hospitalization. The cognitive monitoring tools will be available on a digital platform at no cost to mental health professionals.
“One of the biggest problems in mental healthcare in the United States is that there aren’t enough clinicians and clinical resources to support everyone who needs them,” said Laura Germine, Ph.D., principal investigator on the grant and founding director of the division of brain and cognitive health technology in the Saul Korey Department of Neurology at Einstein. “This grant will allow us to develop tools that can support clinical decision-making and ensure those limited resources get to the right people at the right time.”
Connecting Cognitive Fluctuations and Serious Mental Illness
For people with SMI, evidence suggests that their cognitive abilities—paying attention, remembering new information, and solving problems—may worsen significantly before severe psychiatric episodes occur. For that reason, identifying when patients with SMI are having cognitive problems is a key aspect of the predictive tools that the new project will develop. Other preludes to crises are major fluctuations in symptoms including hallucinations, social withdrawal, apathy, suicidal thoughts, and aggressive behavior.
Dr. Germine has been creating and refining digital tools that incorporate cognitive and behavioral measurements for more than 20 years and will develop and validate the methods needed to detect these changes in cognition and symptoms before the onset of severe episodes.
“To develop tools for predicting when patients need more support, we will conduct a large-scale clinical study to track patients’ changes in cognition, symptoms, and healthcare use over time. Those findings will allow us to develop a learning algorithm that can identify who is at risk. Regular monitoring of patients with SMI will also lead to personalized models that can identify when a specific person is at risk,” Dr. Gemine noted.
Harnessing Technology to Address Challenges
Dr. Germine and colleagues will recruit 1,500 participants receiving inpatient psychiatric care at McLean Hospital in Boston. Over three days, the researchers will conduct a series of brief cognitive assessments throughout the day, evaluate sleep-wake patterns, and review clinical records so they can prospectively predict clinical health outcomes including changes in symptoms, length of hospital stay, and rehospitalization over the next year.
The research team will then follow 250 participants for three months after they leave inpatient psychiatric care, collecting data on daily changes in cognition and symptoms and build personalized risk models. These same digital tools and risk prediction algorithms will then be evaluated among participants with elevated mental health concerns who are patients at Montefiore Health System. “We want to verify that these tools work as well in the Bronx as they do in Boston, even among people who aren’t receiving inpatient psychiatric care or who have varying levels of English fluency.” Dr. Germine said.
“We are committed to ensuring this tool is useful for as many people as possible,” Dr. Germine said. “Especially for those who face the greatest barriers to care and often have the greatest need.”
The grant titled “Dynamic Cognitive Phenotypes for Prediction of Mental Health Outcomes in Serious Mental Illness” (1UF1MH135991) was awarded by the National Institute of Mental Health, part of the NIH.
***
About Albert Einstein College of Medicine
Albert Einstein College of Medicine is one of the nation’s premier centers for research, medical education and clinical investigation. During the 2024-25 academic year, Einstein is home to 712 M.D. students, 226 Ph.D. students, 112 students in the combined M.D./Ph.D. program, and approximately 250 postdoctoral research fellows. The College of Medicine has more than 2,000 full-time faculty members located on the main campus and at its clinical affiliates. In 2024, Einstein received more than $192 million in awards from the National Institutes of Health. This includes the funding of major research centers at Einstein in cancer, aging, intellectual development disorders, diabetes, clinical and translational research, liver disease, and AIDS. Other areas where the College of Medicine is concentrating its efforts include developmental brain research, neuroscience, cardiac disease, and initiatives to reduce and eliminate ethnic and racial health disparities. Its partnership with Montefiore, the University Hospital and academic medical center for Einstein, advances clinical and translational research to accelerate the pace at which new discoveries become the treatments and therapies that benefit patients. For more information, please visit einsteinmed.edu, follow us on Twitter, Facebook, Instagram, LinkedIn, and view us on YouTube.