RTUNet: Residual transformer UNet specifically for pancreas segmentation. (January 2023)

• Record Type: Journal Article
• Authors: Qiu, Chengjian; Liu, Zhe; Song, Yuqing; Yin, Jing; Han, Kai; Zhu, Yan; Liu, Yi; Sheng, Victor S.
• Is Part Of: Biomedical signal processing and control. Volume 79(2023)Part 2
• Publication Date: 2023-01
• Subjects: Pancreas segmentation -- Residual transformer -- Dual convolutional down-sampling -- Dice Hausdorff distance; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.104173
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: Accurate pancreas segmentation is crucial for the diagnostic assessment of pancreatic cancer. However, large position changes, high variability in shape and size, and the extremely blurred boundary make the task of pancreas segmentation challenging. To alleviate these challenges, we propose the residual transformer UNet (RTUNet) to fit the nature of the pancreas. Specifically, a residual transformer block is implemented to extract multi-scale features from a global perspective which captures high variabilities in the pancreas position. In addition, a dual convolutional down-sampling strategy is leveraged to obtain precise shape and size features of the pancreas in a large receptive field which prevents the loss of information. We finally propose a dice hausdorff distance loss that makes the network focus on the pancreas boundary. Through extensive experiments on the public NIH dataset, we achieved a dice similarity coefficient (DSC) of 86.25%, which outperforms the state-of-the-art DSC of 85.49%. In addition, our method surpasses the baselines by more than 3.0% on DSC and improves the min DSC by 2.93%. Furthermore, ablation studies are also performed to prove the effectiveness of each proposed module. Highlights: Residual transformer block can captures the relative pancreas position. Dual down-sampling block addresses inaccurate pancreas morphology caused by pooling. Hausdorff distance constraint makes the network focus on the pancreas boundary. Residual transformer UNet is specifically proposed for pancreas segmentation.