A lightweight classification network for single-lead atrial fibrillation based on depthwise separable convolution and attention mechanism

• Published in: Nan fang yi ke da xue xue bao = Journal of Southern Medical University Vol. 45; no. 3; p. 650
• Main Authors: Hong, Yong, Zhang, Xin, Lin, Mingjun, Wu, Qiucen, Chen, Chaomin
• Format: Journal Article
• Language: Chinese
• Published: China 20.03.2025
• Subjects: Atrial Fibrillation - diagnosis; Deep Learning; Electrocardiography; Humans; Neural Networks, Computer; Wearable Electronic Devices; atrial fibrillation; electrocardiogram; MobileNet; lightweight convolutional neural network; convolutional block attention module
• ISSN: 1673-4254
• DOI: 10.12122/j.issn.1673-4254.2025.03.23

To design a deep learning model that balances model complexity and performance to enable its integration into wearable ECG monitoring devices for automated diagnosis of atrial fibrillation. This study was performed based on data from 84 patients with atrial fibrillation, 25 patients with atrial fibrillation, and 18 subjects without obvious arrhythmia collected from the publicly available datasets LTAFDB, AFDB, and NSRDB, respectively. A lightweight attention network based on depthwise separable convolution and fusion of channel-spatial information, namely DSC-AttNet, was proposed. Depthwise separable convolution was introduced to replace standard convolution and reduce model parameters and computational complexity to realize high efficiency and light weight of the model. The multilayer hybrid attention mechanism was embedded to compute the attentional weights of the channels and spatial information at different scales to improve the feature expression ability of the model. Ten-fold cross-validation was perfor