Related Experiment Video
Updated: Jun 23, 2025
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Aberrant patterns of spontaneous brain activity in schizophrenia: A resting-state fMRI study and classification
Rong Zhang1, Juanjuan Ren1, Xiaoxia Lei1
1Shanghai Key Laboratory of Psychotic Disorders, Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
Background:
Schizophrenia is a prevalent mental disorder, leading to severe disability. Currently, the absence of objective biomarkers hinders effective diagnosis. This study was conducted to explore the aberrant spontaneous brain activity and investigate the potential of abnormal brain indices as diagnostic biomarkers employing machine learning methods.
Methods:
A total of sixty-one schizophrenia patients and seventy demographically matched healthy controls were enrolled in this study. The static indices of resting-state functional magnetic resonance imaging (rs-fMRI) including amplitude of low frequency fluctuations (ALFF), fractional ALFF (fALFF), regional homogeneity (ReHo), and degree centrality (DC) were calculated to evaluate spontaneous brain activity. Subsequently, a sliding-window method was then used to conduct temporal dynamic analysis. The comparison of static and dynamic rs-fMRI indices between the patient and control groups was conducted using a two-sample t-test. Finally, the machine learning analysis was applied to estimate the diagnostic value of abnormal indices of brain activity.
Results:
Schizophrenia patients exhibited a significant increase ALFF value in inferior frontal gyrus, alongside significant decreases in fALFF values observed in left postcentral gyrus and right cerebellum posterior lobe. Pervasive aberrations in ReHo indices were observed among schizophrenia patients, particularly in frontal lobe and cerebellum. A noteworthy reduction in voxel-wise concordance of dynamic indices was observed across gray matter regions encompassing the bilateral frontal, parietal, occipital, temporal, and insular cortices. The classification analysis achieved the highest values for area under curve at 0.87 and accuracy at 81.28% when applying linear support vector machine and leveraging a combination of abnormal static and dynamic indices in the specified brain regions as features.
Conclusions:
The static and dynamic indices of brain activity exhibited as potential neuroimaging biomarkers for the diagnosis of schizophrenia.
More Related Videos
08:36Dynamic Inter-subject Functional Connectivity Reveals Moment-to-Moment Brain Network Configurations Driven by Continuous or Communication Paradigms
Published on: March 21, 2019
13:08Measurement of Fronto-limbic Activity Using an Emotional Oddball Task in Children with Familial High Risk for Schizophrenia
Published on: December 2, 2015