Identify autism spectrum disorder via dynamic filter and deep spatiotemporal feature extraction
作者：
Highlights：
• An effective model is proposed to identify children with autism spectrum disorder.
• Dynamic filters are introduced to customize feature maps for each scanpath.
• Field of view (FoV) maps are proposed to extract fixation-specific features.
• A module based on scanpath and saliency prediction is proposed to generate spatiotemporal features.
摘要
•An effective model is proposed to identify children with autism spectrum disorder.•Dynamic filters are introduced to customize feature maps for each scanpath.•Field of view (FoV) maps are proposed to extract fixation-specific features.•A module based on scanpath and saliency prediction is proposed to generate spatiotemporal features.
论文关键词：Autism spectrum disorder identification,Dynamic filter,Eye movement,Field of view
论文评审过程：Received 10 May 2020, Revised 14 January 2021, Accepted 7 February 2021, Available online 16 February 2021, Version of Record 24 February 2021.
论文官网地址：https://doi.org/10.1016/j.image.2021.116195
