2020/04/21 by Marcel Bengs, Bengs, Marcel, Nils Gessert +3 · 1 citation
Neuroscience · #Autism Spectrum Disorder Research #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Functional Brain Connectivity Studies #Image and Video Processing (eess.IV) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2004.10165
openalex publication_date 2020/04/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Autism spectrum disorder (ASD) is associated with behavioral and\ncommunication problems. Often, functional magnetic resonance imaging (fMRI) is\nused to detect and characterize brain changes related to the disorder.\nRecently, machine learning methods have been employed to reveal new patterns by\ntrying to classify ASD from spatio-temporal fMRI images. Typically, these\nmethods have either focused on temporal or spatial information processing.\nInstead, we propose a 4D spatio-temporal deep learning approach for ASD\nclassification where we jointly learn from spatial and temporal data. We employ\n4D convolutional neural networks and convolutional-recurrent models which\noutperform a previous approach with an F1-score of 0.71 compared to an F1-score\nof 0.65.\n