According to the National Institute of Mental Health, neuropsychiatric disorders are the leading cause of disability in the U.S. Neuropsychiatric conditions include OCD, anxiety, schizophrenia, post-traumatic stress disorder, mood disorders like bipolar disorder, depression, and mania, and degenerative diseases like dementia, Parkinson’s, and Alzheimer’s. The costs associated with these conditions are among the highest of all medical disorders, because they often begin early in life, need ongoing treatment, and can result in years lost to disability. Improving outcomes for people with these conditions is hampered by high costs, late detection, and lack of measurement of contributing factors in a person’s day-to-day environment.

Now Mindstrong Health, a company that works to improve the diagnosis and management of brain disorders, has received funding from Optum Ventures, a $250 million venture fund focused on investing in startup and early-stage companies seeking to advance health care. Mindstrong Health’s technology uses machine learning and artificial intelligence to evaluate smartphone interactions to help diagnose and treat neuropsychiatric and neurodegenerative disorders. The company’s measurement approach is called “digital phenotyping,” which makes assessments based on smartphone use. Digital phenotyping uses three kinds of signals from a smartphone. Sensors measure activity, location, and social meta-data like number of messages going out vs messages coming in. Human-computer interactions capture keyboard performance like typing and clicking. Voice and speech data analyzed with natural language processing yields insights into emotion and cognitive coherence. These signals contribute to the holistic picture of mood, cognition, and behavior called the digital phenotype. Mindstrong says this approach works because smartphones have been widely adopted and provide an opportunity to take measurements passively, objectively, and continuously through the digital phenotyping data collection process. The company used machine learning analytics in three clinical studies with over 200 person-years of data to show that specific digital features correlate with cognitive function, clinical symptoms, and measures of brain activity.

Mindstrong believes its technology will deliver insights that have not been possible with traditional clinical, neuropsychological, and imaging techniques. If technology can deliver an inexpensive way to measure cognition and behavior in any environment with a device most of us use every day, detection will increase, and more people who need help for brain disorders may get it.