基于高效通道注意力的UNet肺结节CT图像分割

doi:10.16088/j.issn.1001-6600.2021071202

广西师范大学学报（自然科学版） ›› 2022, Vol. 40 ›› Issue (3): 66-75.doi: 10.16088/j.issn.1001-6600.2021071202

基于高效通道注意力的UNet肺结节CT图像分割

万黎明¹, 张小乾², 刘知贵², 宋林², 周莹³, 李理^1*

1.西南科技大学计算机科学与技术学院, 四川绵阳 621000;
2.西南科技大学信息工程学院, 四川绵阳 621000;
3.绵阳市中心医院放射科, 四川绵阳 621000

收稿日期:2021-07-12 修回日期:2021-09-09 出版日期:2022-05-25 发布日期:2022-05-27
通讯作者: 李理(1983—), 男, 四川绵阳人, 西南科技大学副教授, 博士。E-mail: ll@swust.edu.cn
基金资助:
国家自然科学基金(61772272, 62102331)

CT Image Segmentation of UNet Pulmonary Nodules Based on Efficient Channel Attention

WAN Liming¹, ZHANG Xiaoqian², LIU Zhigui², SONG Lin², ZHOU Ying³, LI Li^1*

1. School of Computer Science and Technology, Southwest University of Science and Technology, Mianyang Sichuan 621000, China;
2. School of Information Engineering, Southwest University of Science and Technology, Mianyang Sichuan 621000, China;
3. Department of Radiology, Mianyang Central Hospital, Mianyang Sichuan 621000, China

Received:2021-07-12 Revised:2021-09-09 Online:2022-05-25 Published:2022-05-27

摘要/Abstract

摘要： 肺癌是全球死亡率最高的癌症之一,肺结节作为肺癌早期诊断的重要依据,对其进行精准分割格外重要。为了帮助医生诊断肺部病变,本文提出一种改进的UNet肺结节分割方法。首先,在特征提取部分引入高效通道注意力网络(efficient channel attention for deep convolutional neural networks, EcaNet),提高UNet分割效果,使其具有良好的泛化能力。接着,为了降低模型参数量、提升算法分割性能,提出一种基于深度可分离卷积的特征融合模型,用深度可分离卷积代替传统卷积完成特征融合。然后,针对肺结节图像特点,将基于重叠度损失函数(dice loss)与加权交叉熵(weighted cross entropy, WCE)结合作为新的损失函数。最后,为验证所提算法Eca-UNet的有效性,在LIDC-IDRI肺结节公开数据集上进行评估。结果表明:Eca-UNet算法在DICE相似系数、MIOU上比UNet分割算法分别提高10.47、7.34个百分点;同时在训练速度上提升了10.10%,预测速度提升了11.56%。

关键词: 图像分割, 肺结节CT图像, 注意力机制, UNet, 残差网络

Abstract: Lung cancer is one of the cancers with the highest mortality in the world. As an important basis for early diagnosis of lung cancer, accurate segmentation of pulmonary nodules is particularly important. In order to help doctors diagnose lung lesions, an improved UNet lung nodule segmentation method is proposed. First, Efficient Channel Attention for Deep Convolutional Neural Networks(EcaNet) is introduced in the feature extraction part, which improves the UNet segmentation effect and makes it have good generalization ability. At the same time, in order to reduce the number of parameters of the model and improve the segmentation performance of the algorithm, a feature fusion model of depthwise separable convolution is proposed, which replaces the traditional convolution operation with depthwise separable convolution to complete feature fusion. According to the image characteristics of pulmonary nodules, the Dice Loss and the weighted cross entropy (WCE) are combined as a new loss function. To verify the effectiveness of the proposed algorithm Eca-UNet, our evaluated on the LIDC-IDRI public dataset of lung nodules. The results show that the DICE similarity coefficient and MIOU of the Eca-UNet algorithm are 10.47% and 7.34% higher than that of the UNet segmentation algorithm, respectively. At the same time, the training speed has increased by 10.10%, and the prediction speed has increased by 11.56%.

Key words: image segmentation, CT images of pulmonary nodules, attention mechanism, UNet, residual network

中图分类号:

R730.44

万黎明, 张小乾, 刘知贵, 宋林, 周莹, 李理. 基于高效通道注意力的UNet肺结节CT图像分割[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 66-75.

WAN Liming, ZHANG Xiaoqian, LIU Zhigui, SONG Lin, ZHOU Ying, LI Li. CT Image Segmentation of UNet Pulmonary Nodules Based on Efficient Channel Attention[J]. Journal of Guangxi Normal University(Natural Science Edition), 2022, 40(3): 66-75.

参考文献

[1]赫捷, 李霓, 陈万青,等. 中国肺癌筛查与早诊早治指南(2021,北京)[J]. 中华肿瘤杂志, 2021, 43(3):243-268. DOI:10.3760/cma.j.cn112152-20210119-00060.
[2]钟思华,郭兴明,郑伊能. 改进U-Net网络的肺结节分割方法[J]. 计算机工程与应用,2020,56(17):203-209. DOI:10.3778/j.issn.1002-8331.1911-0124.
[3]RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[C]// Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015. Cham: Springer, 2015: 234-241.DOI:10.1007/978-3-319-24574-4_28.
[4]易三莉,陈建亭,贺建峰. ASR-UNet:一种基于注意力机制改进的视网膜血管分割算法[J]. 山东大学学报(理学版),2021,56(9):13-20.DOI:10.6040/j.issn.1671-9352.0.2020.655.
[5]王海鸥,刘慧,郭强,等.面向医学图像分割的超像素U-Net网络设计[J]. 计算机辅助设计与图形学学报,2019,31(6):1007-1017.
[6]ZHANG Y, CHEN K T, CHANG K T, et al. Automatic breast and fibroglandular tissue segmentation in breast MRI using deep learning by a fully-convolutional residual neural network U-Net[J]. Academic Radiology, 2019, 26(11):1526-1535.DOI:10.1016/j.acra.2019.01.012.
[7]王磐,强彦,杨晓棠,等.基于双注意力3D-UNet的肺结节分割网络模型[J].计算机工程,2021,47(2):307-313. DOI:10.19678/j.issn.1000-3428.0057019.
[8]WU W H, GAO L, DUAN H H, et al. Segmentation of pulmonary nodules in CT images based on 3D-UNET combined with three-dimensional conditional random field optimization[J]. Medical Physics, 2020, 47(9): 4054-4063. DOI:10.1002/mp.14248.
[9]XIAO Z T, LIU B W ,GENG L, et al. Segmentation of lung nodules using improved 3D-UNet neural network[J]. Symmetry, 2020, 12(11):1787. DOI:10.3390/sym12111787.
[10]王雪.基于U-Net多尺度和多维度特征融合的皮肤病变分割方法[J].吉林大学学报(理学版),2021,59(1):123-127.DOI:10.13413/j.cnki.jdxblxb.2020178.
[11]伍长荣,接标,叶明全. CT图像肺结节计算机辅助检测与诊断技术研究综述[J].数据采集与处理,2016,31(5):868-881. DOI:10.16337/j.1004-9037.2016.05.003.
[12]HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway: IEEE, 2016: 770-778.DOI: 10.1109/CVPR.2016.90.
[13]邓文轩,杨航,靳婷.基于注意力机制的图像分类降维方法[J].广西师范大学学报(自然科学版),2021,39(2):32-40. DOI: 10.16088/j.issn.1001-6600.2020090704.
[14]WANG Q L, WU B G, ZHU P F, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway:IEEE, 2020:11531-11539.
[15]SANDLER M, HOWARD A, ZHU M L, et al. MobileNetV2: Inverted residuals and linear bottlenecks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE,2018: 4510-4520. DOI:10.1109/CVPR. 2018.00474.
[16]徐学良.人工神经网络的发展及现状[J].微电子学,2017,47(2):239-242. DOI:10.13911/j.cnki.1004-3365.2017.02.022.
[17]LIU Y C, TAN D S, CHEN J C, et al. Segmenting hepatic lesions using residual attention U-Net with an adaptive weighted dice loss[C]// 2019 IEEE Conference on Image Processing(ICIP). Piscataway:IEEE, 2019:3322-3326. DOI:10.1109/ICIP.2019.8803471.
[18]KINGMA D P, BA J.Adam: a method for stochastic optimizatio[EB/OL]. (2015-07-23)[2021-07-12]. https://arxiv.org/abs/1412.6980v8. DOI: 10.48550/arXiv.1412.6980.
[19]ZHOU Z W, SIDDIQUEE M M R, TAJBAKHSH N, et al. UNet++: a nested U-Net architecture for medical image segmentation[C]// Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Cham: Springer, 2018: 3-11.DOI:10.1007/978-3-030-00889-5_1.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于高效通道注意力的UNet肺结节CT图像分割

CT Image Segmentation of UNet Pulmonary Nodules Based on Efficient Channel Attention

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

Metrics

本文评价

推荐阅读 10

[1]	李正光, 陈恒, 林鸿飞. 基于双向语言模型的社交媒体药物不良反应识别[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 40-48.
[2]	张萍, 徐巧枝. 基于多感受野与分组混合注意力机制的肺结节分割研究[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 76-87.
[3]	孔亚钰, 卢玉洁, 孙中天, 肖敬先, 侯昊辰, 陈廷伟. 面向强化当前兴趣的图神经网络推荐算法研究[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 151-160.
[4]	吴军, 欧阳艾嘉, 张琳. 基于多头注意力机制的磷酸化位点预测模型[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 161-171.
[5]	邓文轩, 杨航, 靳婷. 基于注意力机制的图像分类降维方法[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 32-40.
[6]	李维勇, 柳斌, 张伟, 陈云芳. 一种基于深度学习的中文生成式自动摘要方法[J]. 广西师范大学学报（自然科学版）, 2020, 38(2): 51-63.
[7]	王健, 郑七凡, 李超, 石晶. 基于ENCODER_ATT机制的远程监督关系抽取[J]. 广西师范大学学报（自然科学版）, 2019, 37(4): 53-60.
[8]	王勋, 李廷会, 潘骁, 田宇. 基于改进模糊C均值聚类与Otsu的图像分割方法[J]. 广西师范大学学报（自然科学版）, 2019, 37(4): 68-73.
[9]	武文雅, 陈钰枫, 徐金安, 张玉洁. 基于高层语义注意力机制的中文实体关系抽取[J]. 广西师范大学学报（自然科学版）, 2019, 37(1): 32-41.
[10]	岳天驰, 张绍武, 杨亮, 林鸿飞, 于凯. 基于两阶段注意力机制的立场检测方法[J]. 广西师范大学学报（自然科学版）, 2019, 37(1): 42-49.
[11]	夏海英. 基于改进的SLIC区域合并的宫颈细胞图像分割[J]. 广西师范大学学报（自然科学版）, 2016, 34(4): 93-100.
[12]	张新明, 张玉珊, 李振云. 一种改进的矩不变图像分割方法[J]. 广西师范大学学报（自然科学版）, 2011, 29(2): 185-190.
[13]	冯嘉礼, 杨润泽. 属性论方法在图像分割中的应用研究[J]. 广西师范大学学报（自然科学版）, 2011, 29(2): 191-194.