fMRI Detection of Asperger's Disorder using Support Vector Machine Classification

Yash Shailesh Shah¹, Daehyun Yoon¹, Opal Ousley², Xiaoping Hu², Scott J. Peltier¹

Aspergers Disorder is a type of high functioning autism. An earlier study showed that resting state fMRI scans of subjects with Aspergers Disorder show less synchronized activity between nodes of the default mode network. In this study, we have employed a machine learning algorithm using Support Vector Machine(SVM) to classify Aspergers Disorder subjects from normal subjects. It also elicits the possibility of using Support Vector Regression(SVR) to quantify the severity of autism by relating it to the fMRI resting state functional connectivity measures.