News Releases & Research Results Development of a method to distinguish between schizophrenia and developmental disorder based on a machine-learning classification using neuroimaging data

The results of collaborative research and development project conducted by Associate Professor Shinsuke Koike of the Center for Evolutionary Cognitive Sciences, Graduate School of Arts and Sciences, the University of Tokyo; Professor Kiyoto Kasai of the Department of Neuropsychiatry, the University of Tokyo Hospital; Professor Hidenori Yamasue of the Department of Psychiatry, Hamamatsu University School of Medicine (Former Associate Professor of the Department of Neuropsychiatry, the University of Tokyo Hospital); and others.

• Conventionally, a machine-learning classification using neuroimaging data from individuals with psychiatric disorders had mostly been performed to distinguish between typically developing (TD) individuals and those with psychiatric disorders. In this R&D, the research group performed machine learning of neuroimaging data and successfully developed machine-learning classifiers capable of distinguishing between schizophrenia and developmental disorders.
• It was confirmed for the first time that machine-learning classifiers distinguishing between different disorders could be applied to individuals in the early stages of a disorder.
• The machine-learning classifiers developed in this R&D should be applied as markers for differential diagnosis and therapeutic prediction, which are necessary in clinical settings.

This R&D project was conducted with the support of the Strategic International Brain Science Research Promotion Program (Brain/MINDS Beyond) by AMED.

The results of this R&D project were published in Translational Psychiatry on August 17.

Walid Yassin, et al. Machine learning classification using neuroimaging data in schizophrenia, autism, ultra-high risk and first episode psychosis Translational Psychiatry
DOI：10.1038/s41398-020-00965-5