基于拆分残差网络的糖尿病视网膜病变分类

doi:10.16088/j.issn.1001-6600.2023051805

摘要/Abstract

摘要： 糖尿病视网膜病变是一种常见的糖尿病并发症。为提高糖尿病视网膜病变图像分级准确率,本文提出基于拆分残差网络的分级算法。通过融合归一化注意力,增强识别关键特征信息能力,使模型对病灶特征信息提取更具有针对性;利用全局上下文模块综合考虑不同尺度及网络层学习到的特征信息,进一步联系不同时期糖尿病视网膜病灶特点,增强模型表达能力;输出分类器设计多分支结构进行图像分级,提升多类别图像分级精度。实验结果得出模型准确率为94.86%,其他评价指标相比原主干网络模型均有提高。本文模型性能良好,实现了较高精度诊断分级糖尿病视网膜病变图像。

关键词: 医学图像处理, 深度学习, 糖尿病视网膜病变, 注意力机制, 拆分残差网络

Abstract: Diabetic retinopathy is a common complication of diabetes. A classification algorithm based on split residual network is proposed to improve the classification accuracy of diabetic retinopathy images. Through the fusion of normalized attention, the ability to identify key feature information is enhanced, which makes the model more targeted for extracting lesion feature information. The global context module is used to comprehensively consider the feature information learned from different scales and network layers, and further contact the characteristics of diabetic retinopathy in different periods to enhance the expression ability of the model. The output structure uses multi-branch structure to classify images, which avoids the problem that the uneven distribution of data sets affects the classification accuracy. Experimental results show that the model accuracy is 94.86%, and other evaluation indexes are improved compared with the original backbone network model. The proposed model has good performance, and the high-precision diagnosis and grading of diabetic retina images are realized.

Key words: medical image processing, deep learning, diabetic retinopathy, attention mechanism, split residual network

中图分类号: TP391.41

肖宇庭, 吕晓琪, 谷宇, 刘传强. 基于拆分残差网络的糖尿病视网膜病变分类[J]. 广西师范大学学报（自然科学版）, 2024, 42(1): 91-101.

XIAO Yuting, LÜ Xiaoqi, GU Yu, LIU Chuanqiang. Classification of Diabetic Retinopathy Based on Split Residual Network[J]. Journal of Guangxi Normal University(Natural Science Edition), 2024, 42(1): 91-101.

参考文献

[1] 张瑶, 徐前威, 周冉冉, 等. 桃核承气汤辨治糖尿病视网膜病变的机理初探[J]. 中国中医基础医学杂志, 2023, 29(7): 1156-1159. DOI: 10.19945/j.cnki.issn.1006-3250.2023.07.025.
[2] 徐盼盼, 陈长骏, 闫志文, 等. 基于MA-DRNet的糖尿病视网膜病变等级识别方法[J]. 科学技术与工程, 2023, 23(3): 1168-1175. DOI: 10.12404/j.issn.1671-1815.2023.23.03.01168.
[3] DUTTA S, MANIDEEP B C, BASHA S M, et al. Classification of diabetic retinopathy images by using deep learning models[J]. International Journal of Grid and Distributed Computing, 2018, 11(1): 89-106. DOI: 10.14257/ijgdc.2018.11.1.09.
[4] GULSHAN V, PENG L, CORAM M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs[J]. JAMA, 2016, 316(22): 2402-2410. DOI: 10.1001/jama.2016.17216.
[5] LIU Z, MAO H Z, WU C Y, et al. A ConvNet for the 2020s[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2022: 11966-11976. DOI: 10.1109/CVPR52688.2022.01167.
[6] XIE S N, GIRSHICK R, DOLLÁR P, et al. Aggregated residual transformations for deep neural networks[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2017: 5987-5995. DOI: 10.1109/CVPR.2017.634.
[7] BELLO I, FEDUS W, DU X Z, et al. Revisiting ResNets: improved training and scaling strategies[EB/OL]. (2021-03-13)[2023-05-18]. https://arxiv.org/abs/2103.07579. DOI: 10.48550/arXiv.2103.07579.
[8] BELLO I, ZOPH B, VASUDEVAN V, et al. Neural optimizer search with reinforcement learning[EB/OL]. (2017-09-22)[2023-05-18]. https://arxiv.org/abs/1709.07417. DOI: 10.48550/arXiv.1709.07417.
[9] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023. DOI: 10.1109/TPAMI.2019.2913372.
[10] 郑雯, 沈琪浩, 任佳. 基于Improved DR-Net算法的糖尿病视网膜病变识别与分级[J]. 光学学报, 2021, 41(22): 2210002. DOI: 10.3788/AOS202141.2210002.
[11] 顾婷菲, 郝鹏翼, 白琮, 等. 结合多通道注意力的糖尿病性视网膜病变分级[J]. 中国图象图形学报, 2021, 26(7): 1726-1736. DOI: 10.11834/jig.200518.
[12] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Advances in Neural Information Processing Systems 30 (NIPS 2017). Red Hook, NY: Curran Associates Inc., 2017: 6000-6010.
[13] MA Z Q, XIE Q X, XIE P X, et al. HCTNet: a hybrid ConvNet-transformer network for retinal optical coherence tomography image classification[J]. Biosensors, 2022, 12(7): 542. DOI: 10.3390/bios12070542.
[14] ZHANG H, WU C R, ZHANG Z Y, et al. ResNeSt: split-attention networks[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Los Alamitos, CA: IEEE Computer Society, 2022: 2735-2745. DOI: 10.1109/CVPRW56347.2022.00309.
[15] 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. DOI: 10.16088/j.issn.1001-6600.2021071301.
[16] LIU Y C, SHAO Z R, TENG Y Y, et al. NAM: normalization-based attention module[EB/OL]. (2021-11-24)[2023-05-18]. https://arxiv.org/abs/2111.12419. DOI: 10.48550/arXiv.2111.12419.
[17] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[EB/OL]. (2018-07-18)[2023-05-18]. https://arxiv.org/abs/1807.06521. DOI: 10.48550/arXiv.1807.06521.
[18] 谢云霞, 黄海于, 胡建斌. 基于深度卷积神经网络的糖尿病视网膜病变分期及病灶检测[J]. 计算机应用, 2020, 40(8): 2460-2464. DOI: 10.11772/j.issn.1001-9081.2019122198.
[19] IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[EB/OL]. (2015-03-02)[2023-05-18]. https://arxiv.org/abs/1502.03167. DOI: 10.48550/arXiv.1502.03167.
[20] 马青柯. 眼底图像检测和分析系统的设计与开发[D]. 广州: 暨南大学, 2018.
[21] 欧阳继红, 郭泽琪, 刘思光. 糖尿病视网膜病变分期双分支混合注意力决策网络[J]. 吉林大学学报(工学版), 2022, 52(3): 648-656. DOI: 10.13229/j.cnki.jdxbgxb20200813.
[22] LI Y H, YAO T, PAN Y W, et al. Contextual transformer networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(2): 1489-1500. DOI: 10.1109/TPAMI.2022.3164083.
[23] KIM Y, PARK W, ROH M C, et al. GroupFace: learning latent groups and constructing group-based representations for face recognition[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2020: 5620-5629. DOI: 10.1109/CVPR42600.2020.00566.
[24] LI T, GAO Y Q, WANG K, et al. Diagnostic assessment of deep learning algorithms for diabetic retinopathy screening[J]. Information Sciences, 2019, 501: 511-522. DOI: 10.1016/j.ins.2019.06.011.
[25] KHALIFA N E M, LOEY M, TAHA M H N, et al. Deep transfer learning models for medical diabetic retinopathy detection[J]. Acta Informatica Medica, 2019, 27(5): 327-332. DOI: 10.5455/aim.2019.27.327-332.
[26] GU Y, LU X Q, YANG L D, et al. Automatic lung nodule detection using a 3D deep convolutional neural network combined with a multi-scale prediction strategy in chest CTs[J]. Computers in Biology and Medicine, 2018, 103: 220-231. DOI: 10.1016/j.compbiomed.2018.10.011.
[27] 万黎明, 张小乾, 刘知贵, 等. 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. DOI: 10.16088/j.issn.1001-6600.2021071202.
[28] GRAHAM B. Kaggle diabetic retinopathy detection competition report[R/OL]. (2015-08-06)[2023-05-18]. https://storage.googleapis.com/kaggle-forum-message-attachments/88655/2795/competitionreport.pdf.
[29] 纪泽宇, 张兴军, 付哲, 等. 分布式深度学习框架下基于性能感知的DBS-SGD算法[J]. 计算机研究与发展, 2019, 56(11): 2396-2409. DOI: 10.7544/issn1000-1239.2019.20180880.
[30] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2017: 2261-2269. DOI: 10.1109/CVPR.2017.243.
[31] DING X H, ZHANG X Y, MA N N, et al. RepVGG: making VGG-style ConvNets great again[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA: IEEE Computer Society, 2021: 13728-13737. DOI: 10.1109/CVPR46437.2021.01352.
[32] 王明智, 马志强, 赵锋锋, 等. 基于代价敏感正则化和EfficientNet的糖尿病视网膜病变分类方法[J]. 液晶与显示, 2022, 37(12): 1626-1635. DOI: 10.37188/CJLCD.2022-0161.
[33] 宋若仙, 曹鹏, 赵大哲. 结合域适应学习的糖尿病视网膜病变分级诊断[J]. 中国图象图形学报, 2022, 27(11): 3356-3370. DOI: 10.11834/jig.210411.
[34] 程小辉, 李贺军, 邓昀, 等. 基于ME-ANet模型的糖尿病视网膜病变分级[J]. 广西科学, 2022, 29(2): 249-259. DOI: 10.13656/j.cnki.gxkx.20220526.004.
[35] 李家昱, 陈明惠, 杨瑞君, 等. 糖尿病视网膜病变眼底图像筛查研究[J]. 中国激光, 2022, 49(11): 1107001. DOI: 10.3788/CJL202249.1107001.
[36] 梁礼明, 彭仁杰, 冯骏, 等. 基于跨层双线性池化的糖尿病视网膜病变分级算法研究[J]. 生物医学工程学杂志, 2022, 39(5): 928-936. DOI: 10.7507/1001-5515.202104038.
[37] 牛学德, 高丙朋, 任荣荣, 等. 基于轻量级CNN的作物病虫害识别及安卓端应用[J]. 广西师范大学学报(自然科学版), 2022, 40(6): 59-68. DOI: 10.16088/j.issn.1001-6600.2022021601.
[38] 魏明军, 周太宇, 纪占林, 等. 基于YOLOv3的公共场所口罩佩戴检测方法[J]. 广西师范大学学报(自然科学版), 2023, 41(1): 76-86. DOI: 10.16088/j.issn.1001-6600.2022030402.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed