Diagnosis of Schizophrenia and its Subtypes Using MRI and Machine Learning

Motivation: The heterogeneity in schizophrenia remains poorly understood which contributes to the limited success of existing treatments and the observed variability in treatment responses. Goal(s): Our goal was to classify schizophrenia and its subtypes by using machine learning (ML) and MRI to improve understanding of the neurological basis of this schizophrenia. Approach: We applied conventional ML and feature selection methods on MRI to reach our goal. Results: We were able to distinguish schizophrenia and healthy and subtypes of schizophrenia using the combination of MRI and ML. we also showed evidences of brain dysfunctions in schizophrenia and its correlation with behaviors related to the disorderImpact: The outcomes of this study reinforce the notion that the fusion of machine learning methodologies with structural and functional neuroimaging holds the potential to unearth novel biomarkers, consequently contributing to the enhancement of diagnosis and treatment strategies for psychiatric disorders.