Abstract
Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder that can seriously affect children’s physical and mental development. Early diagnosis of autism is crucial for treatment and recovery of normal development. Currently, most diagnoses are based on behavioral observations of symptoms. With the combination of deep learning techniques and medical imaging, MRI-assisted diagnosis methods for autism are increasingly available. The fineness and complexity of ASD lesions bring great challenges to the establishment of robust models. In this paper, a multi-scale 3D-Res2Net-based model with attention subnets for ASD classification is proposed, which can select a suitable fusion strategy for multi-scale information of different layers. The model was trained and tested by fMRI from Autism Brain Imaging Data Exchange dataset. The result of a 5-fold cross-validation shows that the accuracy reaches 75%. Our method has achieved a better classification performance in the assisted diagnosis of autism than conventional deep learning models.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
