[1]顾德,王宁,张寅斌,等.基于深度学习的三维肿瘤及器官分割[J].中国医学物理学杂志,2024,41(9):1122-1128.[doi:DOI:10.3969/j.issn.1005-202X.2024.09.009]
GU De,WANG Ning,ZHANG Yinbin,et al.Three-dimensional tumor and organ segmentation based on deep learning[J].Chinese Journal of Medical Physics,2024,41(9):1122-1128.[doi:DOI:10.3969/j.issn.1005-202X.2024.09.009]
点击复制
基于深度学习的三维肿瘤及器官分割()
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- 卷:
-
41卷
- 期数:
-
2024年第9期
- 页码:
-
1122-1128
- 栏目:
-
医学影像物理
- 出版日期:
-
2024-10-25
文章信息/Info
- Title:
-
Three-dimensional tumor and organ segmentation based on deep learning
- 文章编号:
-
1005-202X(2024)09-1122-07
- 作者:
-
顾德1; 王宁1; 张寅斌2; 刘乐3
-
1.江南大学物联网工程学院, 江苏 无锡 214122; 2.西安交通大学第二附属医院肿瘤科, 陕西 西安 710004; 3.西安交通大学第二附属医院医学影像科, 陕西 西安 710004
- Author(s):
-
GU De1; WANG Ning1; ZHANG Yinbin2; LIU Le3
-
1. School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China 2. Department of Oncology, the Second Affiliated Hospital of Xian Jiaotong University, Xian 710004, China 3. Department of Medical Imaging, the Second Affiliated Hospital of Xian Jiaotong University, Xian 710004, China
-
- 关键词:
-
肿瘤分割; 器官分割; 三维卷积神经网络; 空洞立方集成模块; 跨阶段上下文融合模块
- Keywords:
-
Keywords: tumor segmentation organ segmentation three-dimensional convolutional neural network dilated cubic integration module cross-stage context fusion module
- 分类号:
-
R318;TP183
- DOI:
-
DOI:10.3969/j.issn.1005-202X.2024.09.009
- 文献标志码:
-
A
- 摘要:
-
针对三维医学图像中由于肿瘤或器官的形状、尺度差异较大导致分割精度较低的问题,提出一种端到端的三维全卷积分割模型。首先,设计空洞立方集成模块在不同分辨率阶段实现多尺度集成,增强复杂边界上的识别能力;其次,引入跨阶段上下文融合模块融合浅层和深层特征,促进收敛并更准确地定位目标对象;最后,解码器对来自编码器的特征进行拼接以实现分割。在脑肿瘤分割数据集上,平均Dice相似性系数值达到85.37%;在腹部器官分割数据集上,平均Dice相似性系数值达到83.99%。实验结果表明所提模型在三维肿瘤和器官的分割上具有较高精度。
- Abstract:
-
Abstract: In response to the challenge posed by the significant shape and scale variations of tumors and organs in three-dimensional medical images, which often results in low segmentation accuracy, an end-to-end three-dimensional fully convolutional segmentation model is introduced. A dilated cubic integration module is designed to achieve multi-scale integration at different resolution stages, thereby enhancing the recognition capability on complex boundaries. Subsequently, a cross-stage context fusion module is incorporated to merge shallow and deep features, thereby facilitating convergence and more precise localization of the target objects. Finally, features from the encoder are concatenated by the decoder to realize segmentation. The average Dice similarity coefficients reach 85.37% on the brain tumor segmentation dataset and 83.99% on the abdominal organ segmentation dataset. Experimental results indicate that the proposed model exhibits high accuracy in three-dimensional tumor and organ segmentation.
相似文献/References:
[1]张绿川,杨艳.基于稀疏表示超像素分类的肿瘤超声图像分割算法[J].中国医学物理学杂志,2015,32(06):855.[doi:doi:10.3969/j.issn.1005-202X.2015.06.020]
[J].Chinese Journal of Medical Physics,2015,32(9):855.[doi:doi:10.3969/j.issn.1005-202X.2015.06.020]
备注/Memo
- 备注/Memo:
-
【收稿日期】2024-04-02
【基金项目】江苏省自然科学基金(BK20180594, BK20231036)
【作者简介】顾德,博士,副教授,研究方向:基于图像的生物医学信息识别,E-mail: gude@jiangnan.edu.cn
更新日期/Last Update:
2024-09-26