Identifying Brain Abnormalities with Schizophrenia Based on a Hybrid Feature Selection Technology

Abstract

Many medical imaging data, especially the magnetic resonance imaging (MRI) data, usually have a small sample size, but a large number of features. How to reduce effectively the data dimension and locate accurately the biomarkers from such kinds of data are quite crucial for diagnosis and further precision medicine. In this paper, we propose a hybrid feature selection method based on machine learning and traditional statistical approaches and explore the brain abnormalities of schizophrenia by using the functional and structural MRI data. The results show that the abnormal brain regions are mainly distributed in the supramarginal gyrus, cingulate gyrus, frontal gyrus, precuneus and caudate, and the abnormal functional connections are related to the caudate nucleus, insula and rolandic operculum. In addition, some complex network analyses based on graph theory are utilized on the functional connection data, and the results demonstrate that the located abnormal functional connections in brain can distinguish schizophrenia patients from healthy controls. The identified abnormalities in brain with schizophrenia by the proposed hybrid feature selection method show that there do exist some abnormal brain regions and abnormal disruption of the network segregation and network integration for schizophrenia, and these changes may lead to inaccurate and inefficient information processing and synthesis in the brain, which provide further evidence for the cognitive dysmetria of schizophrenia.

Impact of Psychosocial Occupational Therapy Combined with Anodal Transcranial Direct Current Stimulation to the Left Dorsolateral Prefrontal Cortex on the Cognitive Performance of Patients with Schizophrenia: A Randomized Controlled Trial

“…Chi et al observed that one session of 2 mA anodal and cathodal tDCS on anterior temporal lobes in bilateral design for 13 min can significantly increase the visual memory of healthy people. Although visual memory improved in latter study (Chi et al, 2010), we cannot relate these findings to patients with schizophrenia because there are some identified abnormalities in the brain of patients with schizophrenia (Qiao et al, 2019), furthermore, the stimulation areas were different from the current investigation. Contrary to our results, Gomes et al did not find a significant difference in Visual learning and working memory following 10 sessions of 2 mA tDCS on left DLPFC in patients with schizophrenia (Gomes et al, 2018).…”

Multi-modal biomarkers of low back pain: A machine learning approach

“…We used recursive feature elimination (RFE) ( Guyon et al, 2002 ) in this study. RFE is a popular feature selection approach that is effective in data dimension reduction, increases efficiency of MRI datasets ( Arbabshirani et al, 2017 , Blum and Langley, 1997 , Hall and Smith, 1998 , Kohavi and John, 1997 ), and is applied in many neuroimaging studies ( Qiao et al 2019 ). RFE aids in the elimination of redundant features without incurring substantial loss of information and enables set important features to be used in SVM model training.…”

Machine Learning Analytics of Resting-State Functional Connectivity Predicts Survival Outcomes of Glioblastoma Multiforme Patients

“…First, in the supervised context, as done in this work, predictions were formulated by direct comparisons of abundant functional data and outcome measurements. Recursive feature elimination (RFE) with support vector machines (SVM) were chosen because they outperformed many popular classification algorithms in a survey of neuroimaging studies of brain disorders: simple thresholding, centroid methods, minimum distance, discriminant function analysis, Gaussian process, spectral clustering, fused lasso, random forests, perceptrons, stacked auto-encoder neural networks, SVM without RFE ( 39 , 40 ). Second, appropriately selecting features, which determine the dimensionality of a machine learning model, is critical for SVM in the face of limited outcome data, such as the OS of GBM patients.…”