IBM today announced that, in collaboration with researchers from Harvard Medical School, Mount Sinai School of Medicine, Stanford University, and the Northern California Institute for Research and Education, it will undertake a new initiative funded by the U.S. National Institutes of Health to investigate whether AI can be used to better identify people at risk of developing schizophrenia. It’s a part of the multimillion-dollar, multiyear Accelerating Medicines Partnership (AMP) program, a collaboration between the National Institute of Health (NIH), the U.S. Food and Drug Administration (FDA), pharmaceutical companies, biotech firms, and nonprofit organizations to develop new diagnostics and therapies for patients.
Schizophrenia, a chronic neurological brain disorder estimated to affect over 2.6 million adults in the U.S. alone, often goes undiagnosed. According to the Treatment Advocacy Center, 40% of people with the condition go untreated in any given year. But current medications aren’t a silver bullet, and developing new ones is a costly and time-consuming endeavor. Clinical trials take six to seven years, on average, putting the cost of R&D at roughly $2.6 billion, according to the Pharmaceutical Research and Manufacturers of America.
The NIH’s new Schizophrenia initiative (AMP SZ), which falls within AMP’s purview, is a five-year, $99 million effort to address the heterogeneity of those at risk of developing psychosis, a troublesome symptom of the disease. Alongside IBM, collaborators including the National Alliance on Mental Illness and the American Psychiatric Association Foundation will coordinate activities with the FDA to aggregate, collect, and analyze data related to schizophrenia. Specifically, they’ll work to develop biomarker algorithms from existing data sets and in new data compiled by research investigators within AMP SZ.
Biomarkers can be used to predict a range of health-related outcomes. For instance, they could inform an objective measure of pain and enable remote patient monitoring, freeing up clinicians to address patients who urgently require in-person care.
AMP SZ’s goal is to leverage biomarkers to predict trajectories and outcomes and generate risk calculators that can be used in future trials for treatment intervention, ultimately aiming to halt the progression of psychosis onset, anxiety and mood disorders, alcohol and drug abuse, suicidal ideation and behaviors, and more. IBM says it will contribute its knowledge in data-driven AI and brain imaging for neurodevelopmental and neurodegenerative disorders. In addition, the company plans to survey and guide the collection of data including language samples.
“The … initiative as a whole will be a unique opportunity for IBM to be a leader in the realization the enormous potential of integrating large volumes of data, artificial intelligence, and basic neuroscientific research to help impact mental health,” Guillermo Cecchi, IBM principal research staff member, wrote in a blog post. “The clinical high risk for psychosis syndrome in young people represents an opportune window for early intervention to help prevent the onset of psychosis and other disorders, and to forestall disability.”
IBM’s work in AI schizophrenia diagnostic systems extends back at least as far as 2018, when the company’s psychiatry and neuroimaging groups developed a system capable of predicting, with a claimed over 80% accuracy, the onset of psychosis in a patient. Those researchers built on the findings of a 2015 IBM study demonstrating AI could be used to model differences in speech patterns of patients who later developed psychosis, particularly aberrations in characteristics like syntactic complexity and semantic coherence. Separately, IBM has investigated combining neuroimaging techniques with AI to predict schizophrenia by analyzing a patient’s brain scans.