[1]吕佳,杨鸿强.图结构化辅助监督的视网膜血管分割网络[J].智能系统学报,2026,21(5):1166-1179.[doi:10.11992/tis.202509041]
Lü Jia,YANG Hongqiang.U-Net-based graph-structured auxiliary supervision network for retinal vessel segmentation[J].CAAI transactions on intelligent systems,2026,21(5):1166-1179.[doi:10.11992/tis.202509041]
点击复制
Lü Jia,YANG Hongqiang.U-Net-based graph-structured auxiliary supervision network for retinal vessel segmentation[J].CAAI transactions on intelligent systems,2026,21(5):1166-1179.[doi:10.11992/tis.202509041]
图结构化辅助监督的视网膜血管分割网络
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
21
期数:
2026年第5期
页码:
1166-1179
栏目:
学术论文—机器学习
出版日期:
2026-09-05
- Title:
- U-Net-based graph-structured auxiliary supervision network for retinal vessel segmentation
- Keywords:
- retinal vessel segmentation; deep learning; U-Net; graph-structured auxiliary supervision; vessel skeleton; topological structure; multi-level feature fusion; connectivity
- 分类号:
- TP391.4
- 摘要:
- 现有基于深度学习的视网膜血管分割方法因仅依赖逐像素掩码监督，难以有效建模血管网络的全局连通性和复杂分支结构，常导致分割结果中出现血管断裂等结构性误差。为此，本文提出了一种基于U-Net的图结构化辅助监督视网膜血管分割网络。通过血管骨架引导的结构编码算法，将关键分叉点与连通关系转换为结构化监督信号，引导网络学习视网膜血管的拓扑结构特征。设计了多层次信息流注意力融合模块，有效融合视觉语义特征和拓扑结构特征，提升网络的结构表达能力。利用辅助监督机制，引导网络同时学习拓扑结构特征和视觉语义特征，以兼顾分割精度与血管形态一致性。在CHASEDB1、DRIVE和STARE数据集上的实验结果表明，本文网络在精准分割视网膜血管的同时，可有效减少血管断裂问题，为保障分割结果的生物学一致性提供了参考。
- Abstract:
- Existing deep learning-based retinal vessel segmentation methods, relying exclusively on pixel-wise mask supervision, encounter significant challenges in effectively modeling intricate branching patterns and overall global connectivity features, which leads to structural anomalies such as vascular discontinuities in the segmentation results. Thus, a U-Net-based graph-structured auxiliary supervision network for retinal vessel segmentation is presented in this paper. Firstly, using a vessel skeleton-guided structural encoding algorithm, key bifurcation points and connectivity relationships are transformed into structural supervision cues, enabling the network to capture the global structural attributes of retinal vessels. Secondly, to achieve effective integration of visual-semantic and topological structural features, a multi-level information flow attention fusion module is designed, enhancing the network’s structural representation capability. Finally, an auxiliary supervision mechanism is utilized to guide the network in jointly learning topological structural features and visual-semantic features, balancing segmentation accuracy with vessel morphological consistency. Evaluation on CHASEDB1, DRIVE and STARE datasets reveals that the proposed network achieves accurate retinal vessel segmentation, while effectively mitigates vascular discontinuities issues, thereby providing a reference for preserving biological consistency of segmentation outcomes.
- 参考文献/References:
-
[1] SINGH C. Metabolism and vascular retinopathies: current perspectives and future directions[J]. Diagnostics, 2022, 12(4): 903
[2] LIU Xinghui, TAN Hongwen, WANG Wu, et al. Deep learning based retinal vessel segmentation and hypertensive retinopathy quantification using heterogeneous features cross-attention neural network[J]. Frontiers in medicine, 2024, 11: 1377479
[3] CHEN Shuli, XU Yi, CHEN Bo, et al. Remnant cholesterol is correlated with retinal vascular morphology and diabetic retinopathy in type 2 diabetes mellitus: a cross-sectional study[J]. Lipids in health and disease, 2024, 23(1): 75
[4] SUNG J Y, LEE K H, JUN J H, et al. Changes in peripapillary microvasculature in patients with type 2 diabetes patients: effect of systemic hypertension[J]. Scientific reports, 2023, 13: 19459
[5] 张菡, 贾红艳, 凌赛广, 等. 70岁以上冠心病患者基于人工智能的视网膜血管形态参数研究[J]. 眼科, 2024, 33(4): 285-289 ZHANG Han, JIA Hongyan, LING Saiguang, et al. Correlation analysis between retinal vascular morphology parameters and 70 years-old or over coronary heart disease based on artificial intelligence[J]. Ophthalmology in China, 2024, 33(4): 285-289
[6] 吕佳, 王泽宇. 基于深度学习的视网膜血管分割方法综述[J]. 重庆师范大学学报(自然科学版), 2024, 41(4): 110-125 L? Jia, WANG Zeyu. A survey of retinal vessel segmentation algorithms based on deep learning[J]. Journal of Chongqing Normal University (natural science), 2024, 41(4): 110-125
[7] 汪有崧, 裴峻鹏, 李增辉, 等. 深度学习的视网膜血管分割研究综述[J]. 计算机科学与探索, 2024, 18(8): 1960-1978 WANG Yousong, PEI Junpeng, LI Zenghui, et al. Review of research on deep learning in retinal blood vessel segmentation[J]. Journal of frontiers of computer science and technology, 2024, 18(8): 1960-1978
[8] LIU Li, QIANG Ma, OUYANG Cheng, et al. Topology optimization in medical image segmentation with fast Euler characteristic[EB/OL]. (2025-08-05)[2025-09-27]. https://doi.org/10.48550/arXiv.2507.23763.
[9] RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 234-241.
[10] 缪林一, 李峰. 基于动态特征图卷积网络的视网膜血管分割方法[J]. 中国激光, 2024, 51(15): 181-191 MIAO Linyi, LI Feng. Retinal vessel segmentation based on dynamic feature graph convolutional network[J]. Chinese journal of lasers, 2024, 51(15): 181-191
[11] 吴晨玥, 易本顺, 章云港, 等. 基于改进卷积神经网络的视网膜血管图像分割[J]. 光学学报, 2018, 38(11): 125-131 WU Chenyue, YI Benshun, ZHANG Yungang, et al. Retinal vessel image segmentation based on improved convolutional neural network[J]. Acta optica sinica, 2018, 38(11): 125-131
[12] 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 YI Sanli, CHEN Jianting, HE Jianfeng. ASR-UNet: an improved retinal vessels segmentation algorithm based on attention mechanism[J]. Journal of Shandong University (natural science), 2021, 56(9): 13-20
[13] YI Yugen, GUO Changlu, HU Yangtao, et al. BCR-UNet: bi-directional ConvLSTM residual U-Net for retinal blood vessel segmentation[J]. Frontiers in public health, 2022, 10: 1056226
[14] JIA Wenran, MA Simin, GENG Peng, et al. DT-net: joint dual-input transformer and CNN for retinal vessel segmentation[J]. Computers, materials & continua, 2023, 76(3): 3393-3411.
[15] TONG Le, LI Tianjiu, ZHANG Qian, et al. LiViT-Net: a U-Net-like, lightweight Transformer network for retinal vessel segmentation[J]. Computational and structural biotechnology journal, 2024, 24: 213-224
[16] LI Daxiang, SU Miao, LIU Ying. MSPDD-net: Mamba semantic perception dual decoding network for retinal image vessel segmentation[J]. Computers in biology and medicine, 2025, 193: 110370
[17] 张蓉蓉, 吕晓琪, 李菁, 等. 双路径编码和跨级解耦的视网膜血管分割[J]. 光学精密工程, 2025, 33(14): 2242-2261 ZHANG Rongrong, L? Xiaoqi, LI Jing, et al. Dual-path encoding and cross-level decoupling for retinal vessel segmentation[J]. Optics and precision engineering, 2025, 33(14): 2242-2261
[18] 彭圆圆, 黎浩洋, 李文, 等. MFAL-Mamba: 多尺度特征聚合和注意力引导的眼底血管分割算法[J]. 光学学报, 2025, 45(22): 214-224 PENG Yuanyuan, LI Haoyang, LI Wen, et al. MFAL-mamba: fundus blood vessel segmentation algorithm based on multi-scale features and attention mechanism[J]. Acta optica sinica, 2025, 45(22): 214-224
[19] 王植炜, 黄梓航, 郭英杰, 等. 基于GAN的视网膜血管分割标签优化方法[J]. 华中科技大学学报(自然科学版), 2025, 53(9): 108-113 WANG Zhiwei, HUANG Zihang, GUO Yingjie, et al. GAN-based label refinement method for retinal vessel segmentation[J]. Journal of Huazhong University of Science and Technology (natural science edition), 2025, 53(9): 108-113
[20] HERNANDEZ R J, ROBERTS P A, EL-BOURI W K. Advancing treatment of retinal disease through in silico trials[J]. Progress in biomedical engineering, 2023, 5(2): 022002
[21] WANG Zeyu, JIA L V, LIANG Haocheng. Partial class activation mapping guided graph convolution cascaded U-Net for retinal vessel segmentation[J]. Computers in biology and medicine, 2024, 178: 108736
[22] ZENG Shuang, LEE C H, NNAMDI M C, et al. Novel extraction of discriminative fine-grained feature to improve retinal vessel segmentation[EB/OL]. (2025-05-06)[2025-09-27]. https://doi.org/10.48550/arXiv.2505.03896.
[23] SHIT S, PAETZOLD J C, SEKUBOYINA A, et al. clDice - a novel topology-preserving loss function for tubular structure segmentation[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2021: 16555-16564.
[24] LUX L, BERGER A H, WEERS A, et al. Topograph: an efficient graph-based framework for strictly topology preserving image segmentation[EB/OL]. (2025-04-17)[2025-09-27]. https://doi.org/10.48550/arXiv.2411.03228.
[25] SCHACHT B, GREVING I, FRINTROP S, et al. ContextLoss: context information for topology-preserving segmentation[EB/OL]. (2025-06-10)[2025-09-27]. https://doi.org/10.48550/arXiv.2506.11134.
[26] ZHOU Yi, AHMED T S, WANG Meng, et al. Masked vascular structure segmentation and completion in retinal images[J]. IEEE transactions on medical imaging, 2025, 44(6): 2492-2503
[27] ZHANG T Y, SUEN C Y. A fast parallel algorithm for thinning digital patterns[J]. Communications of the ACM, 1984, 27(3): 236-239
[28] HE Songtao, BASTANI F, JAGWANI S, et al. Sat2Graph: road graph extraction through graph-tensor encoding[C]//Computer Vision–ECCV 2020. Cham: Springer, 2020: 51-67.
[29] AGUIAR M, CASTANO F, TRUJILLO M. Segmentation and detection of vascular bifurcations and crossings in retinal images[C]//Advances in Computing. Cham: Springer, 2018: 432-443.
[30] OWEN C G, RUDNICKA A R, NIGHTINGALE C M, et al. Retinal arteriolar tortuosity and cardiovascular risk factors in a multi-ethnic population study of 10-year-old children; the Child Heart and Health Study in England (CHASE)[J]. Arteriosclerosis, thrombosis, and vascular biology, 2011, 31(8): 1933-1938
[31] STAAL J, ABRAMOFF M D, NIEMEIJER M, et al. Ridge-based vessel segmentation in color images of the retina[J]. IEEE transactions on medical imaging, 2004, 23(4): 501-509
[32] HOOVER A D, KOUZNETSOVA V, GOLDBAUM M. Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response[J]. IEEE transactions on medical imaging, 2000, 19(3): 203-210
[33] HU Xiaoling, LI Fuxin, SAMARAS D, et al. Topology-preserving deep image segmentation[C]//Proceedings of the 33rd international conference on neural information processing systems. Red Hook: Curran Associates Inc. , 2019: 5657-5668.
[34] ZHOU Zongwei, RAHMAN SIDDIQUEE M M, 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.
[35] OKTAY O, SCHLEMPER J, LE FOLGOC L, et al. Attention U-Net: learning where to look for the pancreas[EB/OL]. (2018-05-20)[2025-09-27]. https://doi.org/10.48550/arXiv.1804.03999.
[36] JHA D, SMEDSRUD P H, RIEGLER M A, et al. ResUNet++: an advanced architecture for medical image segmentation[C]//2019 IEEE International Symposium on Multimedia. San Diego: IEEE, 2019: 225-2255.
[37] LI Liangzhi, VERMA M, NAKASHIMA Y, et al. IterNet: retinal image segmentation utilizing structural redundancy in vessel networks[C]//2020 IEEE Winter Conference on Applications of Computer Vision. Snowmass: IEEE, 2020: 3645-3654.
- 相似文献/References:
-
[1]张媛媛,霍静,杨婉琪,等.深度信念网络的二代身份证异构人脸核实算法[J].智能系统学报,2015,10(2):193.[doi:10.3969/j.issn.1673-4785.201405060]
ZHANG Yuanyuan,HUO Jing,YANG Wanqi,et al.A deep belief network-based heterogeneous face verification method for the second-generation identity card[J].CAAI transactions on intelligent systems,2015,10():193.[doi:10.3969/j.issn.1673-4785.201405060]
[2]丁科,谭营.GPU通用计算及其在计算智能领域的应用[J].智能系统学报,2015,10(1):1.[doi:10.3969/j.issn.1673-4785.201403072]
DING Ke,TAN Ying.A review on general purpose computing on GPUs and its applications in computational intelligence[J].CAAI transactions on intelligent systems,2015,10():1.[doi:10.3969/j.issn.1673-4785.201403072]
[3]马晓,张番栋,封举富.基于深度学习特征的稀疏表示的人脸识别方法[J].智能系统学报,2016,11(3):279.[doi:10.11992/tis.201603026]
MA Xiao,ZHANG Fandong,FENG Jufu.Sparse representation via deep learning features based face recognition method[J].CAAI transactions on intelligent systems,2016,11():279.[doi:10.11992/tis.201603026]
[4]刘帅师,程曦,郭文燕,等.深度学习方法研究新进展[J].智能系统学报,2016,11(5):567.[doi:10.11992/tis.201511028]
LIU Shuaishi,CHENG Xi,GUO Wenyan,et al.Progress report on new research in deep learning[J].CAAI transactions on intelligent systems,2016,11():567.[doi:10.11992/tis.201511028]
[5]马世龙,乌尼日其其格,李小平.大数据与深度学习综述[J].智能系统学报,2016,11(6):728.[doi:10.11992/tis.201611021]
MA Shilong,WUNIRI Qiqige,LI Xiaoping.Deep learning with big data: state of the art and development[J].CAAI transactions on intelligent systems,2016,11():728.[doi:10.11992/tis.201611021]
[6]王亚杰,邱虹坤,吴燕燕,等.计算机博弈的研究与发展[J].智能系统学报,2016,11(6):788.[doi:10.11992/tis.201609006]
WANG Yajie,QIU Hongkun,WU Yanyan,et al.Research and development of computer games[J].CAAI transactions on intelligent systems,2016,11():788.[doi:10.11992/tis.201609006]
[7]黄心汉.A3I:21世纪科技之光[J].智能系统学报,2016,11(6):835.[doi:10.11992/tis.201605022]
HUANG Xinhan.A3I: the star of science and technology for the 21st century[J].CAAI transactions on intelligent systems,2016,11():835.[doi:10.11992/tis.201605022]
[8]宋婉茹,赵晴晴,陈昌红,等.行人重识别研究综述[J].智能系统学报,2017,12(6):770.[doi:10.11992/tis.201706084]
SONG Wanru,ZHAO Qingqing,CHEN Changhong,et al.Survey on pedestrian re-identification research[J].CAAI transactions on intelligent systems,2017,12():770.[doi:10.11992/tis.201706084]
[9]杨梦铎,栾咏红,刘文军,等.基于自编码器的特征迁移算法[J].智能系统学报,2017,12(6):894.[doi:10.11992/tis.201706037]
YANG Mengduo,LUAN Yonghong,LIU Wenjun,et al.Feature transfer algorithm based on an auto-encoder[J].CAAI transactions on intelligent systems,2017,12():894.[doi:10.11992/tis.201706037]
[10]王科俊,赵彦东,邢向磊.深度学习在无人驾驶汽车领域应用的研究进展[J].智能系统学报,2018,13(1):55.[doi:10.11992/tis.201609029]
WANG Kejun,ZHAO Yandong,XING Xianglei.Deep learning in driverless vehicles[J].CAAI transactions on intelligent systems,2018,13():55.[doi:10.11992/tis.201609029]
备注/Memo
收稿日期:2025-9-30。
基金项目:国家自然科学基金重大项目(11991024)；重庆市自然科学基金创新发展联合基金重点项目(CSTB2025NSCQ-LZX0074)；重庆市研究生科研创新项目(CYS260405).
作者简介:吕佳，教授，博士，主要研究方向为机器学习、数据挖掘及其在医学图像处理等方面的应用。主持或参与国家、省部级科研项目20余项，发表学术论文70余篇。E-mail：lvjia@cqnu.edu.cn。;杨鸿强，硕士研究生，主要研究方向为深度学习及其在医学图像处理中的应用。E-mail: 2024110516027@stu.cqnu.edu.cn。
通讯作者:吕佳. E-mail：lvjia@cqnu.edu.cn
更新日期/Last Update:
2026-09-05