Detection of ASD Children through Deep-Learning Application of fMRI
Open Access
Abstract
Autism spectrum disorder (ASD) necessitates prompt diagnostic scrutiny to enable immediate, targeted interventions. This study unveils an advanced convolutional-neural-network (CNN) algorithm that was meticulously engineered to examine resting-state functional magnetic resonance imaging (fMRI) for early ASD detection in pediatric cohorts. The CNN architecture amalgamates convolutional, pooling, batch-normalization, dropout, and fully connected layers, optimized for high-dimensional data interpretation. Rigorous preprocessing yielded 22,176 two-dimensional echo planar samples from 126 subjects (56 ASD, 70 controls) who were sourced from the Autism Brain Imaging Data Exchange (ABIDE I) repository. The model, trained on 17,740 samples across 50 epochs, demonstrated unparalleled diagnostic metrics—accuracy of 99.39%, recall of 98.80%, precision of 99.85%, and an F1 score of 99.32%—and thereby eclipsed extant computational methodologies. Feature map analyses substantiated the model’s hierarchical feature extraction capabilities. This research elucidates a deep learning framework for computer-assisted ASD screening via fMRI, with transformative implications for early diagnosis and intervention.Keywords
Funding Information
- Project of Nanjing Rehabilitation Medical Center in Jiangsu (879, 2208085US02, 2016YFCl306205)
- Anhui Province Natural Science Foundation of China (879, 2208085US02, 2016YFCl306205)
- National key research, development plan major chronic non-communicable disease prevention and control research key special sub-project (879, 2208085US02, 2016YFCl306205)
This publication has 46 references indexed in Scilit:
- Neural mechanisms of cue-approach trainingNeuroImage, 2017
- The importance of temporal dynamics in the transition from suicidal thought to behavior.Clinical Psychology: Science and Practice, 2016
- Neuronal mechanisms and circuits underlying repetitive behaviors in mouse models of autism spectrum disorderBehavioral and Brain Functions, 2016
- Going deeper with convolutions2015
- Performance-optimized hierarchical models predict neural responses in higher visual cortexProceedings of the National Academy of Sciences, 2014
- The autism brain imaging data exchange: towards a large-scale evaluation of the intrinsic brain architecture in autismMolecular Psychiatry, 2013
- Functional connectivity magnetic resonance imaging classification of autismBrain, 2011
- Neural signatures of autismProceedings of the National Academy of Sciences, 2010
- Precursors and prodromata of schizophrenia: findings from the Edinburgh High Risk Study and their literature contextPsychological Medicine, 2006
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998