基于混合注意力U-net全脑全脊髓临床靶区自动勾画  

Automatic delineation of craniospinal clinical target volume based on hybrid attention U-net

作　　者：李红伟 倪春霞 陈淑 孟歌 胡小洋 汪洋 Li Hongwei;Ni Chunxia;Chen Shu;Meng Ge;Hu Xiaoyang;Wang Yang(Department of Radiation Oncology,Shanghai Gamma Hospital,Shanghai 200235,China;Radiation Oncology Center,Huashan Hospital,Fudan University,Shanghai 200052,China)

机构地区：[1]上海伽玛医院放疗科,上海200235 [2]复旦大学附属华山医院放射治疗中心,上海200052

出　　处：《中华放射肿瘤学杂志》2022年第3期266-271,共6页Chinese Journal of Radiation Oncology

基　　金：国家自然科学基金项目(11775098)。

摘　　要：目的在U-net卷积神经网络基础上设计出混合注意力U-net(HA-U-net)网络用于全脑全脊髓临床靶体积(CTV)自动勾画,并与U-net自动分割模型分割结果进行比较。方法研究回顾了110例全脑全脊髓患者数据,选择80例用于训练集,10例用于验证集,20例作为测试集。HA-U-net以U-net为基准网络,在U-net网络输入端加入双注意力模块,同时在跳跃连接中加入注意力门模块来建立全脑全脊髓CTV自动勾画网络模型。评估参数为戴斯相似性系数、豪斯多夫距离和精确率。结果HA-U-net网络得到戴斯相似性系数为0.901±0.041,豪斯多夫距离为(2.77±0.29)mm,精确率为0.903±0.038,结果均优于U-net网络分割结果(均P<0.05)。结论HA-U-net卷积神经网络可以有效提升全脑全脊髓CTV自动分割的精度,有助于医生提高工作效率与勾画一致性。Objective Hybrid attention U-net(HA-U-net)neural network was designed based on U-net for automatic delineation of craniospinal clinical target volume(CTV)and the segmentation results were compared with those of U-net automatic segmentation model.Methods The data of 110 craniospinal patients were reviewed,Among them,80 cases were selected for the training set,10 cases for the validation set and 20 cases for the test set.HA-U-net took U-net as the basic network architecture,double attention module was added at the input of U-net network,and attention gate module was combined in skip-connection to establish the craniospinal automatic delineation network model.The evaluation parameters included Dice similarity coefficient(DSC),Hausdorff distance(HD)and precision.Results The DSC,HD and precision of HA-U-net network were 0.901±0.041,2.77±0.29 mm and 0.903±0.038,respectively,which were better than those of U-net(all P<0.05).Conclusion The results show that HA-U-net convolutional neural network can effectively improve the accuracy of automatic segmentation of craniospinal CTV,and help doctors to improve the work efficiency and the consistent delineation of CTV.

关 键 词：深度学习 卷积神经网络 自动分割 全脑全脊髓临床靶体积 

分 类 号：R730.55[医药卫生—肿瘤] TP18[医药卫生—临床医学] TP391.41[自动化与计算机技术—控制理论与控制工程]