A multi-view convolutional neural network method combining attention mechanism for diagnosing autism spectrum disorder

• Published in: PloS one Vol. 18; no. 12; p. e0295621
• Main Authors: Wang, Mingzhi, Ma, Zhiqiang, Wang, Yongjie, Liu, Jing, Guo, Jifeng
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 08.12.2023; Public Library of Science (PLoS)
• Subjects: Accuracy; Algorithms; Analysis; Artificial neural networks; Attention; Autism; Brain; Brain mapping; Care and treatment; Data integration; Datasets; Diagnosis; Diagnostic imaging; Diagnostic systems; Functional magnetic resonance imaging; Health aspects; Interpersonal communication in children; Learning; Machine learning; Magnetic resonance; Magnetic resonance imaging; Medical diagnosis; Medical imaging; Methods; Neural networks; Neurodevelopmental disorders; Neuroimaging; Pervasive developmental disorders; Psychological aspects; Supervised learning; Unsupervised learning; China
• DOI: 10.1371/journal.pone.0295621
• ISSN: 1932-6203

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition whose current psychiatric diagnostic process is subjective and behavior-based. In contrast, functional magnetic resonance imaging (fMRI) can objectively measure brain activity and is useful for identifying brain disorders. However, the ASD diagnostic models employed to date have not reached satisfactory levels of accuracy. This study proposes the use of MAACNN, a method that utilizes multi-view convolutional neural networks (CNNs) in conjunction with attention mechanisms for identifying ASD in multi-scale fMRI. The proposed algorithm effectively combines unsupervised and supervised learning. In the initial stage, we employ stacked denoising autoencoders, an unsupervised learning method for feature extraction, which provides different nodes to adapt to multi-scale data. In the subsequent stage, we perform supervised learning by employing multi-view CNNs for classification and obtain the final results. Finally, multi-scale data fusion is achieved by using the attention fusion mechanism. The ABIDE dataset is used to evaluate the model we proposed., and the experimental results show that MAACNN achieves superior performance with 75.12% accuracy and 0.79 AUC on ABIDE-I, and 72.88% accuracy and 0.76 AUC on ABIDE-II. The proposed method significantly contributes to the clinical diagnosis of ASD.