• 1. School of Information Science and Engineering, Yunnan University, Kunming 650504, P. R. China;
  • 2. Third Affiliated Hospital of Kunming Medical University, Kunming 650118, P. R. China;
LANG Xun, Email: langxun@ynu.edu.cn
Export PDF Favorites Scan Get Citation

Existing classification methods for myositis ultrasound images have problems of poor classification performance or high computational cost. Motivated by this difficulty, a lightweight neural network based on a soft threshold attention mechanism is proposed to cater for a better IIMs classification. The proposed network was constructed by alternately using depthwise separable convolution (DSC) and conventional convolution (CConv). Moreover, a soft threshold attention mechanism was leveraged to enhance the extraction capabilities of key features. Compared with the current dual-branch feature fusion myositis classification network with the highest classification accuracy, the classification accuracy of the network proposed in this paper increased by 5.9%, reaching 96.1%, and its computational complexity was only 0.25% of the existing method. The obtained results support that the proposed method can provide physicians with more accurate classification results at a lower computational cost, thereby greatly assisting them in their clinical diagnosis.

Citation: TAN Hao, LANG Xun, WANG Tao, HE Bingbing, LI Zhiyao, LU Yu, ZHANG Yufeng. A lightweight convolutional neural network for myositis classification from muscle ultrasound images. Journal of Biomedical Engineering, 2024, 41(5): 895-902. doi: 10.7507/1001-5515.202301023 Copy

  • Previous Article

    Recurrence prediction of gastric cancer based on multi-resolution feature fusion and context information
  • Next Article

    Feature detection of B-ultrasound images of intussusception in children based on improved YOLOv8n