ASD children adaption behaviour assessment via hand movement properties: a RoadMap

Dinghuang Zhang, Carrie Miriam Toptan, Shuwen Zhao, Gongyue Zhang, Honghai Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Adaptive behavioural assessments are useful in the diagnosis of autism. This research proposes a strategy of assessing autistic children’s adaption skills through the change of hand behaviour complexity based on deep learning and complex systems. Specifically, we implement a sparse representation of high-dimensional features of hand movements utilize convolutional neural network (CNN) and Bag of Word model (BoW) and explain in detail how two quantify measurements (complexity and diversity) reflect adaption behavioural capacity. This paper introduces our ongoing projects and demonstrates the preliminary experimental setups and motion protocol design. Future work includes improving the interaction scenarios, establishing a data set, and enhancing the interpretability of the results of the adaption behaviour skill measurements.
Original languageEnglish
Title of host publicationAdvances in Computational Intelligence Systems
Subtitle of host publicationContributions Presented at the 20th UK Workshop on Computational Intelligence, September 8-10, 2021, Aberystwyth, Wales, UK
EditorsThomas Jansen, Richard Jensen, Neil Mac Parthaláin, Chih-Min Lin
PublisherSpringer
Pages475–480
Volume1409
ISBN (Electronic)9783030870942
DOIs
Publication statusPublished - 18 Nov 2021

Publication series

NameAdvances in Intelligent Systems and Computing
PublisherSpringer
Volume1409
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365

Keywords

  • ASD diagnose
  • Adaption skill assessment
  • CNN
  • Computer vision

Fingerprint

Dive into the research topics of 'ASD children adaption behaviour assessment via hand movement properties: a RoadMap'. Together they form a unique fingerprint.

Cite this