Method comparisons for differentiation of Schizophrenia and Bipolar based on rs-fMRI Intrinsic and Functional Networks

The study compared various brain imaging methods (rs-fMRI) to differentiate between schizophrenia and bipolar disorder, two psychiatric illnesses with overlapping symptoms. The research involved 371 patients with psychosis and was validated on a separate group of 315 patients. Researchers tested multiple approaches including neural networks and traditional machine learning models to analyze functional brain networks. The best results were achieved using temporal profiles of intrinsic brain networks. Static functional connectivity measurements performed better than more complex dynamic methods. Although the overall accuracy of distinguishing between these two disorders remained modest, the research highlights the stability of simpler analysis methods and emphasizes the challenges in finding reliable biomarkers for psychiatric disorders.