Digital Phenotyping: Data-Driven Clinical Signatures for People with Hallucinations

This multi-year project is funded by a $12.6 Million grant award from the National Institute of Mental Health (NIMH). The study is a collaboration between the University of Washington, University of Minnesota, and Louisiana State. Our multidisciplinary team is collecting data from a large sample of people who experience hallucinations in all 50 U.S. States using an array of smartphone behavioral measurement tools: audio diaries, self-reports, and performance-based cognitive tests. With the aid of artificial intelligence and novel computational modelling strategies, our team aims to derive data-driven clinical signatures to predict individual differences in severe negative outcomes; identify and mitigate bias in modelling across groups defined by race, sex, and age; examine whether adding smartphone- captured behavioral data to information that is typically available in the clinical record improves model clinical utility; and produce machine learning-ready data structures that adhere to FAIR (Findable, Accessible, Interoperable, Reusable) principles. If successful, these measures and models can be used to guide scalable clinical decision making, resource allocation, treatment, and impactful prevention efforts.

Learn more here: Project Description