In July 2017, the University of Alberta and IBM launched a cooperative initiative to test the potential for artificial intelligence (AI) in diagnosing schizophrenia. They announced that researchers had developed an AI system that used brain scans to diagnose schizophrenia with 74% accuracy. The AI system was also able to predict the severity of specific schizophrenia symptoms, such as attention problems, bizarre behaviors, positive formal thought disorder, lack of motivation or apathy, and the negative symptom of alogia (minimal use of words).
These findings were reported in “Learning stable and predictive network-based patterns of schizophrenia and its clinical symptoms . . .