Assessing visual motor performance in autistic children based on Kinect and fNIRS: A case study

Abstract

In recent years, the incidence rate of children with autism has shown a significant upward trend. Rehabilitation training is an important part of recovery or improvement in autism children. However, during autism rehabilitation training, the methods that can visually reflect and objectively evaluate its effects are seldom considered. Therefore, this study aimed to objectively evaluate the rehabilitation impact of visual-motor skills training in children with autism via quantitative measures. In this study, vision sensors and functional near-infrared spectroscopy were used to monitor and analyze visual motor training task of 20 autism children. These children were divided into high- and low-score groups according to the autism behavior checklist (ABC). Results showed significant differences between the high- and low-score groups in the brain regions of the left and right temporal lobe, right motor cortex, and left occipital lobe; the difference in functional connectivity was greatest when the left hand was moving at the green light (p < 0.05). The differences in speed, acceleration, and angle between the high- and low-score groups were mainly reflected in left-hand movement. Moreover, analysis of multimodal data showed that visual motor training had a positive effect on brain activation and functional connectivity, and increasing the frequency of left-hand training and using more green light were beneficial to the improvement of brain function. These findings can be used as basis to help optimize rehabilitation programs and improve rehabilitation effectiveness.

Keywords: Autistic children; Kinect; Multimodal data; Visual motor; fNIRS.