基于PCU-Net网络的肺肿瘤分割  

Segmentation of lung tumors based on PCU-Net

作　　者：蔡浩 李朋 宫晓梅[2] 王娆芬[1] CAI Hao;LI Peng;GONG Xiaomei;WANG Raofen(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China;Department of Radiation Oncology,Shanghai Pulmonary Hospital,Shanghai 200433,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620 [2]上海市肺科医院放疗科,上海200433

出　　处：《上海工程技术大学学报》2024年第4期444-450,共7页Journal of Shanghai University of Engineering Science

基　　金：国家自然科学基金资助(62173222);上海市科委科技创新行动计划资助(20Y11913600);申康三年行动计划肺科培育项目资助(SKPY2021006)。

摘　　要：深度学习技术可辅助医生进行肿瘤的精准分割。但肺肿瘤与周围组织界限不清楚,现有方法存在分割边缘模糊、模型参数量大等问题。提出一种对轻量级肺肿瘤分割的部分卷积坐标注意力U-net(partial convolution coordinate attention U-net,PCU-Net)算法。引入部分卷积降低模型参数量,同时提升模型特征提取的能力。在U-Net跳跃链接处添加坐标注意力模块,使网络更精准获取肿瘤的位置信息,提高分割精度。研究结果表明,改进的PCU-Net在参数量减少58.57%的同时,Dice值、IoU和Recall分别提高4.22%、4.26%和6.82%。将PCU-Net模型与其他语义分割模型对比显示,PCU-Net的Dice值比其他模型高出3~6百分点。Deep learning techniques can assist doctors in precise tumor segmentation.However,existing methods often suffer from issues such as fuzzy segmentation edges and large model parameter counts due to the unclear boundaries between lung tumors and surrounding tissues.A partial convolution coordinate attention U-net(PCU-Net)algorithm for lightweight lung tumor segmentation was proposed.The partial convolution was introduced to reduce model parameters and enhance feature extraction capability.The coordinate attention module was added at skip connection of PCU-Net,so that more precise localization of tumors was achieved by network and segmentation accuracy was improved.The research result shows that the improved PCU-Net can reduce model parameters by 58.57%while increase Dice coefficient,Intersection over Union(IoU)and Recall by 4.22%,4.26%and 6.82%,respectively.The comparison between PPU-Net and other semantic segmentation models shows that Dice coefficient of PCU-Net is 3-6 percentage points higher than that of other models.

关 键 词：肺肿瘤分割 部分卷积 坐标注意力机制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]