期刊
RADIOLOGICAL PHYSICS AND TECHNOLOGY
卷 14, 期 3, 页码 238-247出版社
SPRINGER JAPAN KK
DOI: 10.1007/s12194-021-00620-8
关键词
Whole breast irradiation; Deep learning; 3D-UNet; Segmentation; Grad-CAM
资金
- Multidisciplinary Cooperative Research Program of the Center for Computational Sciences, University of Tsukuba
This study implemented 3D-CNN for CTV segmentation in whole breast irradiation, achieving DSC >= 0.85 on left, right, and both-sided breast cancer datasets with 3D-UNet. Grad-CAM heatmaps indicated the focus of 3D-UNet in decision-making, highlighting the regions from target-side breast tissue to the opposite-side breast.
This study aims to implement three-dimensional convolutional neural networks (3D-CNN) for clinical target volume (CTV) segmentation for whole breast irradiation and investigate the focus of 3D-CNNs during decision-making using gradient-weighted class activation mapping (Grad-CAM). A 3D-UNet CNN was adopted to conduct automatic segmentation of the CTV for breast cancer. The 3D-UNet was trained using three datasets of left-, right-, and both left- and right-sided breast cancer patients. Segmentation accuracy was evaluated using the Dice similarity coefficient (DSC). Grad-CAM was applied to trained CNNs. The DSCs for the datasets of the left-, right-, and both left- and right-sided breasts were on an average 0.88, 0.89, and 0.85, respectively. The Grad-CAM heatmaps showed that the 3D-UNet used for segmentation determined the CTV region from the target-side breast tissue and by referring to the opposite-side breast. Although the size of the dataset was limited, DSC >= 0.85 was achieved for the segmentation of breast CTV using the 3D-UNet. Grad-CAM indicates the applicable scope and limitations of using a CNN by indicating the focus of such networks during decision-making.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据