Regions of interest computed by SVM wrapped method for Alzheimer’s disease examination from segmented MRI

Abstract

Accurate identification of the most relevant brain regions linked to Alzheimer’s disease (AD) is crucial in order to improve diagnosis techniques and to better understand this neurodegenerative process. For this purpose, statistical classification is suitable. In this work, a novel method based on support vector machine recursive feature elimination (SVM-RFE) is proposed to be applied on segmented brain MRI for detecting the most discriminant AD regions of interest (ROIs). The analyses are performed both on gray and white matter tissues, achieving up to 100% accuracy after classification and outperforming the results obtained by the standard t-test feature selection. The present method, applied on different subject sets, permits automatically determining high-resolution areas surrounding the hippocampal area without needing to divide the brain images according to any common template.

Magnetic resonance imaging biomarkers for the early diagnosis of Alzheimer's disease: a machine learning approach

“…Due to their multivariate properties, machine-learning techniques are able to automatically extract multiple information from image sets without requiring a priori hypotheses of where this information may be coded in the images. Several studies have assessed the diagnostic value of these techniques in the classification of AD by cerebral MRI studies (Davatzikos et al, 2008 ; Klöppel et al, 2008 ; Gerardin et al, 2009 ; Cuingnet et al, 2011 ; Hidalgo-Muñoz et al, 2014 ), showing promising results also for the prediction of conversion in the early stages of disease (Tufail et al, 2012 ; Moradi et al, 2015 ). Among these studies, Klöppel et al ( 2008 ) used machine learning classification and structural MR images for the extraction of spatially-distributed multivariate diagnostic biomarkers.…”

Brain MR Image Classification for Alzheimer’s Disease Diagnosis Based on Multifeature Fusion

“…In the classification based on neuroimaging, several feature selection techniques have been proposed, for example, univariate methods (e.g., t -test) [25], multivariate approaches (e.g., sparse logistic regression) [26], perturbation method [27], and support vector machine recursive feature elimination (SVM-RFE) [28]. SVM-RFE has been successfully implemented in various neuroscience applications [29, 30], but it does not have a good performance on image analysis [31]. In this paper, we improve the process of feature selection by combining the SVM-RFE and covariance.…”

Classification of Alzheimer disease among susceptible brain regions

“…In a study by Davat, several regions such as the HP, amygdala, EC as well as temporal lobe grey matter (GM), insular cortex, posterior cingulate and precuneus, and orbitofrontal cortex were used. Another study of Hidalgo‐Muñoz et al, which is set out to measure the regions of interest (ROIs) computed by SVM‐RFE, correspond intimately to the regions pointed in the literature as most affected by AD, and HP, EC, and parahippocampal region as referred by Du et al and Velayudhan et al, the insular cortex by Xie et al, and among other regions like the amygdala, lenticular nucleus, or fusiform gyrus by Tzourio‐Mazoyer et al The HP is a basic subcortical structure involved in declarative memory consolidation and spatial orientation by Squire and Tsien et al, which are the skills severely affected by AD. In another article, Yong et al used a standard ROI method to analyze the volumes of the HP and EC.…”