An efficient movement and mental classification for children with autism based on motion and EEG features

作者：Dang, Xin^*; Wei, Ran; Li, Guohui

来源：Journal of Ambient Intelligence and Humanized Computing, 2017, 8(6): 907-912.

DOI：10.1007/s12652-016-0424-x

• Autism spectrum disorder (ASD)
• Motion-based game
• Movement detection
• EEG

The early intervention services for the children with autism are labor costly. To settle this problem, some activities classification based rehabilitation training system is developed. However, the mental characteristics of the children with autism are less considered. In this paper, a novel activities classification framework is proposed for the rehabilitation training of the children with autism. The movement and mental based features are obtained using the motion and electroencephalography vectors from Kinect and MindWave. Two support vector machines are integrated to perform a frame-based classification procedure. The proposed methods are tested in a dataset includes 110 recordings from five children (two with autism). The proposed algorithm achieved an accuracy of 96.2, 93.3, 94.06 and 97.40% for the good positive, NG positive, good negative, and NG negative movements, respectively.

• 出版日期2017-11
• 单位天津工业大学