A Hybrid CNN-Transformer for Focal Liver Lesion Classification
Ling Zhao, Shuaiqi Liu, Bing Li, Wenjia Cai, Ping Liang, Jie Yu, Jie Zhao
Abstract
The early diagnosis of focal liver lesions (FLLs) plays a key role in the successful treatment of liver cancer. To effectively diagnose focal liver lesions, we used contrast-enhanced ultrasound (CEUS) to diagnose FLLs. A hybrid CNN and Transformer network is used to extract local and global spatio-temporal features of CEUS. Firstly, the R(2+1)D with pre-trained weights is used to extract local multi-scale spatio-temporal features, and then the feature maps with various size are input into the G- <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> Transformer for global information learning and interaction. To reduce the parameters of the traditional Transformer, a new efficient Transformer named G-Transformer is proposed to achieve better performance with lower parameters. The proposed model was evaluated on a multi-center-multi-disease dataset, and the results showed that the proposed model can achieve an AUC of 0.8237 and better generalization performance.