Effects of Brain Atlases and Machine Learning Methods on the Discrimination of Schizophrenia Patients: A Multimodal MRI Study.

Recently, machine learning techniques have been widely applied in discriminative studies of schizophrenia (SZ) patients with multimodal magnetic resonance imaging (MRI); however, the effects of brain atlases and machine learning methods remain largely unknown. In this study, we collected MRI data for 61 first-episode SZ patients (FESZ), 79 chronic SZ patients (CSZ) and 205 normal controls (NC) and calculated 4 MRI measurements, including regional gray matter volume (GMV), regional homogeneity (ReHo), amplitude of low-frequency fluctuation and degree centrality. We systematically analyzed the performance of two classifications (SZ vs NC; FESZ vs CSZ) based on the combinations of three brain atlases, five classifiers, two cross validation methods and 3 dimensionality reduction algorithms. Our results showed that the groupwise whole-brain atlas with 268 ROIs outperformed the other two brain atlases. In addition, the leave-one-out cross validation was the best cross validation method to select the best hyperparameter set, but the classification performances by different classifiers and dimensionality reduction algorithms were quite similar. Importantly, the contributions of input features to both classifications were higher with the GMV and ReHo features of brain regions in the prefrontal and temporal gyri. Furthermore, an ensemble learning method was performed to establish an integrated model, in which classification performance was improved. Taken together, these findings indicated the effects of these factors in constructing effective classifiers for psychiatric diseases and showed that the integrated model has the potential to improve the clinical diagnosis and treatment evaluation of SZ.

The gut microbiome is associated with brain structure and function in schizophrenia.

The effect of the gut microbiome on the central nervous system and its possible role in mental disorders have received increasing attention. However, knowledge about the relationship between the gut microbiome and brain structure and function is still very limited. Here, we used 16S rRNA sequencing with structural magnetic resonance imaging (sMRI) and resting-state functional (rs-fMRI) to investigate differences in fecal microbiota between 38 patients with schizophrenia (SZ) and 38 demographically matched normal controls (NCs) and explored whether such differences were associated with brain structure and function. At the genus level, we found that the relative abundance of Ruminococcus and Roseburia was significantly lower, whereas the abundance of Veillonella was significantly higher in SZ patients than in NCs. Additionally, the analysis of MRI data revealed that several brain regions showed significantly lower gray matter volume (GMV) and regional homogeneity (ReHo) but significantly higher amplitude of low-frequency fluctuation in SZ patients than in NCs. Moreover, the alpha diversity of the gut microbiota showed a strong linear relationship with the values of both GMV and ReHo. In SZ patients, the ReHo indexes in the right STC (r = - 0.35, p = 0.031, FDR corrected p = 0.039), the left cuneus (r = - 0.33, p = 0.044, FDR corrected p = 0.053) and the right MTC (r = - 0.34, p = 0.03, FDR corrected p = 0.052) were negatively correlated with the abundance of the genus Roseburia. Our results suggest that the potential role of the gut microbiome in SZ is related to alterations in brain structure and function. This study provides insights into the underlying neuropathology of SZ.

Divergent Alterations of Structural-Functional Connectivity Couplings in First-episode and Chronic Schizophrenia Patients.

Emerging evidence suggests that the coupling relating the structural connectivity (SC) of the brain to its functional connectivity (FC) exhibits remarkable changes during development, normal aging, and diseases. Although altered structural-functional connectivity couplings (SC-FC couplings) have been previously reported in schizophrenia patients, the alterations in SC-FC couplings of different illness stages of schizophrenia (SZ) remain largely unknown. In this study, we collected structural and resting-state functional MRI data from 73 normal controls (NCs), 61 first-episode (FeSZ) and 78 chronic (CSZ) schizophrenia patients. Positive and negative syndrome scale (PANSS) scores were assessed for all patients. Structural and functional brain networks were constructed using gray matter volume (GMV) and resting-state magnetic resonance imaging (rs-fMRI) time series measurements. At the connectivity level, the CSZ patients showed significantly increased SC-FC coupling strength compared with the FeSZ patients. At the node strength level, significant decreased SC-FC coupling strength was observed in the FeSZ patients compared to that of the NCs, and the coupling strength was positively correlated with negative PANSS scores. These results demonstrated divergent alterations of SC-FC couplings in FeSZ and CSZ patients. Our findings provide new insight into the neuropathological mechanisms underlying the developmental course of SZ.

NEURO-LEARN: a Solution for Collaborative Pattern Analysis of Neuroimaging Data.

The development of neuroimaging instrumentation has boosted neuroscience researches. Consequently, both the fineness and the cost of data acquisition have profoundly increased, leading to the main bottleneck of this field: limited sample size and high dimensionality of neuroimaging data. Therefore, the emphasis of ideas of data pooling and research collaboration has increased over the past decade. Collaborative analysis techniques emerge as the idea developed. In this paper, we present NEURO-LEARN, a solution for collaborative pattern analysis of neuroimaging data. Its collaboration scheme consists of four parts: projects, data, analysis, and reports. While data preparation workflows defined in projects reduce the high dimensionality of neuroimaging data by collaborative computation, pooling of derived data and sharing of pattern analysis workflows along with generated reports on the Web enlarge the sample size and ensure the reliability and reproducibility of pattern analysis. Incorporating this scheme, NEURO-LEARN provides an easy-to-use Web application that allows users from different sites to share projects and processed data, perform pattern analysis, and obtain result reports. We anticipate that this solution will help neuroscientists to enlarge sample size, conquer the curse of dimensionality and conduct reproducible studies on neuroimaging data with efficiency and validity.