BackgroundA lack of a sufficiently large sample at single sites causes poor generalizability in automatic diagnosis classification of heterogeneous psychiatric disorders such as schizophrenia based on brain imaging scans. Advanced deep learning methods may be capable of learning subtle hidden patterns from high dimensional imaging data, overcome potential site-related variation, and achieve reproducible cross-site classification. However, deep learning-based cross-site transfer classification, despite less imaging site-specificity and more generalizability of diagnostic models, has not been investigated in schizophrenia.MethodsA large multi-site functional MRI sample (n = 734, including 357 schizophrenic patients from seven imaging resources) was collected, and a deep discriminant autoencoder network, aimed at learning imaging site-shared functional connectivity features, was developed to discriminate schizophrenic individuals from healthy controls.FindingsAccuracies of approximately 85·0% and 81·0% were obtained in multi-site pooling classification and leave-site-out transfer classification, respectively. The learned functional connectivity features revealed dysregulation of the cortical-striatal-cerebellar circuit in schizophrenia, and the most discriminating functional connections were primarily located within and across the default, salience, and control networks.InterpretationThe findings imply that dysfunctional integration of the cortical-striatal-cerebellar circuit across the default, salience, and control networks may play an important role in the “disconnectivity” model underlying the pathophysiology of schizophrenia. The proposed discriminant deep learning method may be capable of learning reliable connectome patterns and help in understanding the pathophysiology and achieving accurate prediction of schizophrenia across multiple independent imaging sites.

Parkinson's disease (PD) is a neurodegenerative disorder with dysfunction in cortices as well as white matter (WM) tracts. While the changes to WM structure have been extensively investigated in PD, the nature of the functional changes to WM remains unknown. In this study, the regional activity and functional connectivity of WM were compared between PD patients (n = 57) and matched healthy controls (n = 52), based on multimodel magnetic resonance imaging data sets. By tract‐based spatial statistical analyses of regional activity, patients showed decreased structural‐functional coupling in the left corticospinal tract compared to controls. This tract also displayed abnormally increased functional connectivity within the left post‐central gyrus and left putamen in PD patients. At the network level, the WM functional network showed small‐worldness in both controls and PD patients, yet it was abnormally increased in the latter group. Based on the features of the WM functional connectome, previously un‐evaluated individuals could be classified with fair accuracy (73%) and area under the curve of the receiver operating characteristics (75%). These neuroimaging findings provide direct evidence for WM functional changes in PD, which is crucial to understand the functional role of fiber tracts in the pathology of neural circuits.

This study identifies several novel susceptibility genes for MMD. The association with homocysteine metabolism and the immune system enrichment of susceptibility gene expression suggest that therapeutic interventions targeting these pathways may be effective approaches for MMD treatment.

No abstract