计算机工程与科学2025，Vol.47Issue(4)：677-685,9.DOI:10.3969/j.issn.1007-130X.2025.04.011
CT图像肾肿瘤分割的三维轴向Transformer模型
3D axial Transformer model for kidney tumor segmentation in CT images
张金龙 1吴敏 2孙玉宝1
作者信息
- 1. 南京信息工程大学计算机学院、网络空间安全学院,江苏南京 210044
- 2. 东部战区总医院医学工程科,江苏南京 210018
- 折叠
摘要
Abstract
Automatic segmentation of kidneys and their tumor areas in CT image sequences can pro-vide quantitative references for radiotherapy and chemotherapy planning.Currently,kidney tumor segmentation models based on Transformer have attracted widespread attention,especially when used in conjunction with the U-Net model and its variants.Existing Transformer-based segmentation networks typically learn features within local windows of individual slices,resulting in insufficient representation zof intra-slice spatial information and inter-slice axial information.To address this issue,a three-dimensional axial Transformer module is proposed,which decomposes the complex coupling of the three dimensions into alternating axial attentions,integrating both intra-slice and inter-slice axial correlation information.Based on the three-dimensional axial Transformer module,a two-stage kidney tumor seg-mentation encoder-decoder network,ATrans UNet(Axial Transformer UNet),incorporates multi-scale features and residual learning.On KiTS19 dataset,the Dice similarity coefficients for kidney and kidney tumor segmentation are 96.43％and 81.04％,respectively,representing an improvement of 8.40％over 2D-Unet and 4.84％over 3D-Unet in average Dice scores.关键词
CT图像序列/肾肿瘤三维分割/三维轴向Transformer/二阶段编解码网络Key words
CT image sequences/3D segmentation of kidney tumors/3D axial Transformer/two-stage encoding-decoding network分类
信息技术与安全科学引用本文复制引用
张金龙,吴敏,孙玉宝..CT图像肾肿瘤分割的三维轴向Transformer模型[J].计算机工程与科学,2025,47(4):677-685,9.
