吉首大学学报（自然科学版）2026，Vol.47Issue(2)：34-43,10.DOI:10.13438/j.cnki.jdzk.2026.02.006
基于改进的SAMUS的甲状腺结节分割
Thyroid Nodule Segmentation Based on Improved SAMUS
罗翔 1谭子涵 1谢剑宇 1李曙2
作者信息
- 1. 吉首大学通信与电子工程学院,湖南 吉首 416000
- 2. 广州医科大学生物医学工程学院,广东 广州 511000
- 折叠
摘要
Abstract
To address the issues of noise interference,low contrast,and blurred boundaries in thyroid nod-ule ultrasound images,an improved SAMUS was constructed based on the SAMUS framework.This model employs a shallow feature extraction module to obtain multi-scale detail information and intro-duces a shallow feature fusion strategy in the decoding stage to supplement shallow representations and enhance adaptability to morphological changes of nodules.A comparative experiment conducted between the improved SAMUS and four traditional models shows that compared with the U-Net,U-Net v2,DAE-Former and SAMUS,the improved SAMUS can more accurately restore the target shape,effectively re-duce mis-segmentation and under-segmentation phenomena;compared with the SAMUS,the IOU of the improved SAMUS increases by 0.89,the Dice increases by 0.73,and the HD95 decreases by 0.30.This indicates that the improved SAMUS can obtain shallow multi-scale detailed features and fuse them dur-ing the decoding stage,which can improve the overlap degree of thyroid nodule image segmentation and reduce boundary errors.关键词
甲状腺结节/超声图像分割/多尺度融合/浅层特征提取/SAMUS/深度学习Key words
thyroid nodules/ultrasound image segmentation/multi-scale fusion/shallow feature extrac-tion/SAMUS/deep learning分类
信息技术与安全科学引用本文复制引用
罗翔,谭子涵,谢剑宇,李曙..基于改进的SAMUS的甲状腺结节分割[J].吉首大学学报（自然科学版）,2026,47(2):34-43,10.
