Paper Info
Title
The Genetic-Evolutionary Random Support Vector Machine Cluster Analysis In Autism Spectrum Disorder
Abstract
Previous researches have produced a number of conclusions on the functional magnetic resonance imaging (fMRI) study for autism spectrum disorder (ASD) patients, but there are different opinions about the brain regions of the lesions. In order to study ASD more deeply, an advanced framework, i.e., genetic-evolutionary random support vector machine (SVM) cluster, was proposed in this paper. In our method, an initial cluster of multiple SVMs was first built by randomly picking samples and features. Then, these SVMs were selected to recombine and mutate the aim of genetic evolution until the number of genetic evolution which reached the threshold or the classification accuracy was stable. We evaluated the proposed method by using the resting state fMRI data (103 ASD patients and 106 healthy controls), which achieved a 96.8% accuracy. Based on the classification results, the abnormal brain regions were found out. This study suggests the pathogenesis of ASD to a certain extent and offers great assistance for the diagnosis of potential patients with ASD.
Year
DOI
Venue
2019
10.1109/ACCESS.2019.2902889
IEEE ACCESS
Keywords
Field
DocType
Genetic-evolutionary random SVM cluster, fMRI, graph metrics, autism spectrum disorder, abnormal brain regions
Autism,Genetic Evolution,Pattern recognition,Functional magnetic resonance imaging,Computer science,Resting state fMRI,Support vector machine,Feature extraction,Artificial intelligence,Autism spectrum disorder,Distributed computing
Journal
Volume
ISSN
Citations 
7
2169-3536
0
PageRank 
References 
Authors
0.34
0
7
Name
Order
Citations
PageRank
XiaAn Bi133.10
Yingchao Liu231.07
qi sun34612.34
Xianhao Luo400.34
Haiyan Tan500.34
Jie Chen69138.15
Nianyin Zeng727821.91