Recognition of early-onset schizophrenia using deep-learning method

Functional connectivity at resting state was found altered in early-onset schizophrenia patients. However, its potential as biomarker of clinical diagnosis is unknown. To test whether resting-state functional connectivity can be a potential biomarker of classifying patients from controls, 39 early-onset schizophrenia patients and 31 healthy controls were included in our study. Resting-state functional MRI networks were built with the whole brain atlas as classification features to distinguish patients from healthy controls. Three-stage deep-learning network was used to deduce dimension reduction, and feedforward back propagation neural networks were used as classifier. As the result, the classification accuracy reached 79.3% (87.4% for sensitivity, 82.2% for specificity, p < 0.05 for permuted test). Our works showed us that resting-state connectivity presented good potential classification capacity and can be used as biomarker of clinical diagnosis.

[1]  Daoqiang Zhang,et al.  Ensemble sparse classification of Alzheimer's disease , 2012, NeuroImage.

[2]  M. Fox,et al.  Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging , 2007, Nature Reviews Neuroscience.

[3]  H. Hwu,et al.  Frequency‐specific alternations in the amplitude of low‐frequency fluctuations in schizophrenia , 2014, Human brain mapping.

[4]  Thomas Efferth,et al.  Synergy and Antagonism of Active Constituents of ADAPT-232 on Transcriptional Level of Metabolic Regulation of Isolated Neuroglial Cells , 2013, Front. Neurosci..

[5]  Vince D. Calhoun,et al.  Classification of schizophrenia patients based on resting-state functional network connectivity , 2013, Front. Neurosci..

[6]  K. Vogeley,et al.  Resting-state functional network correlates of psychotic symptoms in schizophrenia , 2010, Schizophrenia Research.

[7]  M. Matějka,et al.  Connectivity of the anterior insula differentiates participants with first-episode schizophrenia spectrum disorders from controls: a machine-learning study , 2016, Psychological Medicine.

[8]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[9]  T. McGlashan,et al.  Early detection and intervention of schizophrenia: rationale and research , 1998, British Journal of Psychiatry.

[10]  S. Heckers,et al.  Functional resting-state networks are differentially affected in schizophrenia , 2011, Schizophrenia Research.

[11]  Abraham Z. Snyder,et al.  Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion , 2012, NeuroImage.

[12]  Dewen Hu,et al.  Discriminative analysis of resting-state functional connectivity patterns of schizophrenia using low dimensional embedding of fMRI , 2010, NeuroImage.

[13]  Karl J. Friston,et al.  Statistical parametric maps in functional imaging: A general linear approach , 1994 .

[14]  N. Volkow,et al.  Aging and Functional Brain Networks , 2011, Molecular Psychiatry.

[15]  Thomas G. Dietterich.  Machine-Learning Research , 1997, AI Mag..

[16]  Lars T. Westlye,et al.  Consistent Functional Connectivity Alterations in Schizophrenia Spectrum Disorder: A Multisite Study , 2017, Schizophrenia bulletin.

[17]  Huafu Chen,et al.  Specific frequency bands of amplitude low‐frequency oscillation encodes personality , 2014, Human brain mapping.

[18]  V. Calhoun,et al.  Classifying Schizophrenia Using Multimodal Multivariate Pattern Recognition Analysis: Evaluating the Impact of Individual Clinical Profiles on the Neurodiagnostic Performance. , 2016, Schizophrenia bulletin.

[19]  J. Bedwell,et al.  Accelerated age-related decline of visual information processing in first-degree relatives of persons with schizophrenia , 2004, Psychiatry Research.

[20]  Alan C. Evans,et al.  Early brain development in infants at high risk for autism spectrum disorder , 2017, Nature.

[21]  B. Biswal,et al.  The resting brain: unconstrained yet reliable. , 2009, Cerebral cortex.

[22]  Martin Wiesmann,et al.  Eyes open and eyes closed as rest conditions: impact on brain activation patterns , 2004, NeuroImage.

[23]  Bharat B. Biswal,et al.  Functional Integration Between Brain Regions at Rest Occurs in Multiple-Frequency Bands , 2015, Brain Connect..

[24]  Hilleke E. Hulshoff Pol,et al.  Classification of schizophrenia patients and healthy controls from structural MRI scans in two large independent samples , 2012, NeuroImage.

[25]  J. Gabrieli,et al.  Hyperactivity and hyperconnectivity of the default network in schizophrenia and in first-degree relatives of persons with schizophrenia , 2009, Proceedings of the National Academy of Sciences.

[26]  Yajing Pang,et al.  Extraversion modulates functional connectivity hubs of resting‐state brain networks , 2016, Journal of neuropsychology.

[27]  Huafu Chen,et al.  Alteration of functional connectivity in autism spectrum disorder: effect of age and anatomical distance , 2016, Scientific Reports.

[28]  V. Calhoun,et al.  High Classification Accuracy for Schizophrenia with Rest and Task fMRI Data , 2012, Front. Hum. Neurosci..

[29]  Edward T. Bullmore,et al.  Schizophrenia, neuroimaging and connectomics , 2012, NeuroImage.

[30]  J. McGrath,et al.  Schizophrenia: a concise overview of incidence, prevalence, and mortality. , 2008, Epidemiologic reviews.

[31]  Andrew B. Leber,et al.  Human Neuroscience Original Research Article Individual Differences in Distraction by Motion Predicted by Neural Activity in Mt/v5 , 2022 .