{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,7]],"date-time":"2026-03-07T19:11:01Z","timestamp":1772910661255,"version":"3.50.1"},"reference-count":36,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:00:00Z","timestamp":1750291200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:00:00Z","timestamp":1750291200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100031032","name":"Fatih Sultan Mehmet Vak\u0131f University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100031032","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Health Inf Sci Syst"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Accurate segmentation of retinal vessels is crucial for the early diagnosis and management of various ocular diseases. Existing methods often struggle to segment thin vessels, leading to missed diagnoses and inaccurate treatment plans. This study proposes a novel Multi-Teacher Based Knowledge Distillation (MTKD) method for Retinal Vessel Segmentation (RVS) to address this challenge. Our approach utilizes the expertise of multiple teacher networks, each specialized in learning different vessel characteristics. Specifically, we train three distinct teacher networks: one on the original ground truth, one on a modified ground truth highlighting thin vessels, and another on a modified ground truth emphasizing thick vessels. The student network is then trained to minimize the knowledge discrepancy between its predictions and the soft predictions of all three teachers. By incorporating knowledge from these specialized teachers, the student network effectively learns to segment both thin and thick vessels with improved accuracy. We evaluate our method on two retinal fundus image datasets and two angiography datasets, demonstrating highly competitive performance compared to state-of-the-art methods. The proposed method improves the baseline U-Net model by up to 8.44 points in F1 and 10.42 points in IOU. Additionally, we introduce a penalization technique to the student model\u2019s loss function, further enhancing segmentation performance. Comprehensive ablation studies validate the effectiveness of the multi-teacher approach, the choice of loss functions, and the impact of model complexity. Our findings suggest that MTKD offers a promising approach for enhancing the robustness and accuracy of RVS. All source code, datasets, and results are made publicly available to support reproducibility and further research.<\/jats:p>","DOI":"10.1007\/s13755-025-00356-4","type":"journal-article","created":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T14:37:23Z","timestamp":1750343843000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Multi-teacher based knowledge distillation for retinal vessel segmentation"],"prefix":"10.1007","volume":"13","author":[{"given":"Abdullah","family":"Eid","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5825-2230","authenticated-orcid":false,"given":"Musa","family":"Aydin","sequence":"additional","affiliation":[]},{"given":"Zeki","family":"Ku\u015f","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,6,19]]},"reference":[{"issue":"10","key":"356_CR1","doi-asserted-by":"publisher","first-page":"1292","DOI":"10.1007\/s10439-022-03058-0","volume":"50","author":"A Khandouzi","year":"2022","unstructured":"Khandouzi A, Ariafar A, Mashayekhpour Z, Pazira M, Baleghi Y. Retinal vessel segmentation, a review of classic and deep methods. Ann Biomed Eng. 2022;50(10):1292\u2013314. https:\/\/doi.org\/10.1007\/s10439-022-03058-0.","journal-title":"Ann Biomed Eng"},{"key":"356_CR2","doi-asserted-by":"publisher","unstructured":"Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation, pp. 234\u2013241. Springer, 2015. https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"356_CR3","doi-asserted-by":"crossref","unstructured":"Wang B, Qiu S, He H. Dual encoding u-net for retinal vessel segmentation. In: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13\u201317, 2019, Proceedings, Part I 22, pp. 84\u201392 2019. Springer.","DOI":"10.1007\/978-3-030-32239-7_10"},{"key":"356_CR4","doi-asserted-by":"crossref","unstructured":"Li D, Dharmawan DA, Ng BP, Rahardja S. Residual u-net for retinal vessel segmentation. In: 2019 IEEE International Conference on Image Processing (ICIP), 2019;1425\u20131429. IEEE.","DOI":"10.1109\/ICIP.2019.8803101"},{"key":"356_CR5","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102025","volume":"70","author":"H Wu","year":"2021","unstructured":"Wu H, Wang W, Zhong J, Lei B, Wen Z, Qin J. Scs-net: a scale and context sensitive network for retinal vessel segmentation. Med Image Anal. 2021;70: 102025.","journal-title":"Med Image Anal"},{"key":"356_CR6","doi-asserted-by":"crossref","unstructured":"Zhou Y, Yu H, Shi H. Study group learning: Improving retinal vessel segmentation trained with noisy labels. In: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2021: 24th International Conference, Strasbourg, France, September 27\u2013October 1, 2021, Proceedings, Part I 24, pp. 57\u201367 2021. Springer.","DOI":"10.1007\/978-3-030-87193-2_6"},{"issue":"9","key":"356_CR7","doi-asserted-by":"publisher","first-page":"4623","DOI":"10.1109\/JBHI.2022.3188710","volume":"26","author":"W Liu","year":"2022","unstructured":"Liu W, Yang H, Tian T, Cao Z, Pan X, Xu W, Jin Y, Gao F. Full-resolution network and dual-threshold iteration for retinal vessel and coronary angiograph segmentation. IEEE J Biomed Health Inform. 2022;26(9):4623\u201334.","journal-title":"IEEE J Biomed Health Inform"},{"key":"356_CR8","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2023.104574","volume":"82","author":"K Sun","year":"2023","unstructured":"Sun K, Chen Y, Chao Y, Geng J, Chen Y. A retinal vessel segmentation method based improved u-net model. Biomed Signal Process Control. 2023;82: 104574. https:\/\/doi.org\/10.1016\/j.bspc.2023.104574.","journal-title":"Biomed Signal Process Control"},{"key":"356_CR9","doi-asserted-by":"crossref","unstructured":"Zhou Z, Rahman&nbsp;Siddiquee MM, Tajbakhsh N, Liang J. Unet++: a nested u-net architecture for medical image segmentation. In: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4, pp. 3\u201311 2018. Springer.","DOI":"10.1007\/978-3-030-00889-5_1"},{"key":"356_CR10","unstructured":"Oktay O, Schlemper J, Folgoc LL, Lee M, Heinrich M, Misawa K, Mori K, McDonagh S, Hammerla NY, Kainz B, et al. Attention u-net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03999 2018."},{"key":"356_CR11","doi-asserted-by":"crossref","unstructured":"Zhang S, Fu H, Yan Y, Zhang Y, Wu Q, Yang M, Tan M, Xu Y. Attention guided network for retinal image segmentation. In: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13\u201317, 2019, Proceedings, Part I 22, 2019;797\u2013805. Springer.","DOI":"10.1007\/978-3-030-32239-7_88"},{"key":"356_CR12","doi-asserted-by":"crossref","unstructured":"Mou L, Zhao Y, Chen L, Cheng J, Gu Z, Hao H, Qi H, Zheng Y, Frangi A, Liu J. Cs-net: Channel and spatial attention network for curvilinear structure segmentation. In: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13\u201317, 2019, Proceedings, Part I 22, 2019;721\u2013730. Springer.","DOI":"10.1007\/978-3-030-32239-7_80"},{"key":"356_CR13","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101556","volume":"58","author":"SY Shin","year":"2019","unstructured":"Shin SY, Lee S, Yun ID, Lee KM. Deep vessel segmentation by learning graphical connectivity. Med Image Anal. 2019;58: 101556. https:\/\/doi.org\/10.1016\/j.media.2019.101556.","journal-title":"Med Image Anal"},{"issue":"4","key":"356_CR14","doi-asserted-by":"publisher","first-page":"1427","DOI":"10.1109\/JBHI.2018.2872813","volume":"23","author":"Z Yan","year":"2019","unstructured":"Yan Z, Yang X, Cheng K-T. A three-stage deep learning model for accurate retinal vessel segmentation. IEEE J Biomed Health Inform. 2019;23(4):1427\u201336. https:\/\/doi.org\/10.1109\/JBHI.2018.2872813.","journal-title":"IEEE J Biomed Health Inform"},{"key":"356_CR15","doi-asserted-by":"publisher","unstructured":"Wu Y, Xia Y, Song Y, Zhang D, Liu D, Zhang C, Cai W. Vessel-net: retinal vessel segmentation under multi-path supervision, pp. 264\u2013272. Springer, 2019. https:\/\/doi.org\/10.1007\/978-3-030-32239-7_30 .","DOI":"10.1007\/978-3-030-32239-7_30"},{"issue":"12","key":"356_CR16","doi-asserted-by":"publisher","first-page":"3384","DOI":"10.1109\/JBHI.2020.3002985","volume":"24","author":"D Wang","year":"2020","unstructured":"Wang D, Haytham A, Pottenburgh J, Saeedi O, Tao Y. Hard attention net for automatic retinal vessel segmentation. IEEE J Biomed Health Inform. 2020;24(12):3384\u201396. https:\/\/doi.org\/10.1109\/JBHI.2020.3002985.","journal-title":"IEEE J Biomed Health Inform"},{"issue":"7","key":"356_CR17","doi-asserted-by":"publisher","first-page":"3314","DOI":"10.1109\/JBHI.2023.3272342","volume":"27","author":"Y Zhi","year":"2023","unstructured":"Zhi Y, Hau WK, Zhang H, Gao Z. Vessel contour detection in intracoronary images via bilateral cross-domain adaptation. IEEE J Biomed Health Inform. 2023;27(7):3314\u201325. https:\/\/doi.org\/10.1109\/JBHI.2023.3272342.","journal-title":"IEEE J Biomed Health Inform"},{"issue":"6","key":"356_CR18","doi-asserted-by":"publisher","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","volume":"129","author":"J Gou","year":"2021","unstructured":"Gou J, Yu B, Maybank SJ, Tao D. Knowledge distillation: a survey. Int J Comput Vision. 2021;129(6):1789\u2013819. https:\/\/doi.org\/10.1007\/s11263-021-01453-z.","journal-title":"Int J Comput Vision"},{"key":"356_CR19","doi-asserted-by":"crossref","unstructured":"Yang C, Yu X, An Z, Xu Y. Categories of response-based, feature-based, and relation-based knowledge distillation. In: Advancements in knowledge distillation: towards new horizons of intelligent systems, pp. 1\u201332. Springer, 2023.","DOI":"10.1007\/978-3-031-32095-8_1"},{"key":"356_CR20","unstructured":"Phuong M, Lampert C. Towards understanding knowledge distillation. In: International Conference on Machine Learning, 2019;5142\u20135151. PMLR."},{"key":"356_CR21","doi-asserted-by":"crossref","unstructured":"Cho JH, Hariharan B. On the efficacy of knowledge distillation. In: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2019;4794\u20134802.","DOI":"10.1109\/ICCV.2019.00489"},{"key":"356_CR22","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2024.111854","volume":"297","author":"S Ahmad","year":"2024","unstructured":"Ahmad S, Ullah Z, Gwak J. Multi-teacher cross-modal distillation with cooperative deep supervision fusion learning for unimodal segmentation. Knowl-Based Syst. 2024;297: 111854. https:\/\/doi.org\/10.1016\/j.knosys.2024.111854.","journal-title":"Knowl-Based Syst"},{"key":"356_CR23","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.124734","volume":"255","author":"Y Qiu","year":"2024","unstructured":"Qiu Y, Lu H, Mei J, Bao S, Xu J. Towards semi-supervised multi-modal rectal cancer segmentation: a large-scale dataset and a multi-teacher uncertainty-aware network. Expert Syst Appl. 2024;255: 124734. https:\/\/doi.org\/10.1016\/j.eswa.2024.124734.","journal-title":"Expert Syst Appl"},{"issue":"12","key":"356_CR24","doi-asserted-by":"publisher","first-page":"15027","DOI":"10.1007\/s10489-022-04252-2","volume":"53","author":"J Gu","year":"2022","unstructured":"Gu J, Tian F, Oh I-S. Retinal vessel segmentation based on self-distillation and implicit neural representation. Appl Intell. 2022;53(12):15027\u201344. https:\/\/doi.org\/10.1007\/s10489-022-04252-2.","journal-title":"Appl Intell"},{"issue":"2","key":"356_CR25","doi-asserted-by":"publisher","first-page":"1937","DOI":"10.1007\/s10489-024-05274-8","volume":"54","author":"Y Wang","year":"2024","unstructured":"Wang Y, Hou Q, Cao P, Yang J, R. Zaiane O. Lesion-aware knowledge distillation for diabetic retinopathy lesion segmentation. Appl Intell. 2024;54(2):1937\u201356. https:\/\/doi.org\/10.1007\/s10489-024-05274-8.","journal-title":"Appl Intell"},{"key":"356_CR26","doi-asserted-by":"crossref","unstructured":"Peng L, Lin L, Cheng P, He H, Tang X. Student becomes decathlon master in retinal vessel segmentation via dual-teacher multi-target domain adaptation. In: International Workshop on Machine Learning in Medical Imaging, 2022;32\u201342. Springer.","DOI":"10.1007\/978-3-031-21014-3_4"},{"issue":"4","key":"356_CR27","doi-asserted-by":"publisher","first-page":"742","DOI":"10.54287\/gujsa.1575986","volume":"11","author":"Z Ku\u015f","year":"2024","unstructured":"Ku\u015f Z. Uknow-net: knowledge-enhanced u-net for improved retinal vessel segmentation. Gazi Univ J Sci Part A. 2024;11(4):742\u201358. https:\/\/doi.org\/10.54287\/gujsa.1575986.","journal-title":"Gazi Univ J Sci Part A"},{"key":"356_CR28","doi-asserted-by":"publisher","unstructured":"Hinton G, Vinyals O, Dean J. Distilling the knowledge in a neural network. arXiv 2015. https:\/\/doi.org\/10.48550\/ARXIV.1503.02531 . https:\/\/arxiv.org\/abs\/1503.02531.","DOI":"10.48550\/ARXIV.1503.02531"},{"key":"356_CR29","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1016\/j.neunet.2023.04.015","volume":"164","author":"R Shang","year":"2023","unstructured":"Shang R, Li W, Zhu S, Jiao L, Li Y. Multi-teacher knowledge distillation based on joint guidance of probe and adaptive corrector. Neural Netw. 2023;164:345\u201356.","journal-title":"Neural Netw"},{"issue":"4","key":"356_CR30","doi-asserted-by":"publisher","first-page":"501","DOI":"10.1109\/TMI.2004.825627","volume":"23","author":"J Staal","year":"2004","unstructured":"Staal J, Abr\u00e0moff MD, Niemeijer M, Viergever MA, Van Ginneken B. Ridge-based vessel segmentation in color images of the retina. IEEE Trans Med Imaging. 2004;23(4):501\u20139.","journal-title":"IEEE Trans Med Imaging"},{"key":"356_CR31","doi-asserted-by":"publisher","first-page":"5895","DOI":"10.1109\/JBHI.2023.3314981","volume":"27","author":"Z Ku\u015f","year":"2023","unstructured":"Ku\u015f Z, Kiraz B. Evolutionary architecture optimization for retinal vessel segmentation. IEEE J Biomed Health Inf. 2023;27:5895\u2013903.","journal-title":"IEEE J Biomed Health Inf"},{"key":"356_CR32","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.bspc.2018.06.007","volume":"46","author":"A Carballal","year":"2018","unstructured":"Carballal A, Novoa FJ, Fernandez-Lozano C, Garc\u00eda-Guimaraes M, Aldama-L\u00f3pez G, Calvi\u00f1o-Santos R, Vazquez-Rodriguez JM, Pazos A. Automatic multiscale vascular image segmentation algorithm for coronary angiography. Biomed Signal Process Control. 2018;46:1\u20139.","journal-title":"Biomed Signal Process Control"},{"key":"356_CR33","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2020.105769","volume":"198","author":"PM Samuel","year":"2021","unstructured":"Samuel PM, Veeramalai T. Vssc net: vessel specific skip chain convolutional network for blood vessel segmentation. Comput Methods Programs Biomed. 2021;198: 105769.","journal-title":"Comput Methods Programs Biomed"},{"issue":"24","key":"356_CR34","doi-asserted-by":"publisher","first-page":"5507","DOI":"10.3390\/app9245507","volume":"9","author":"F Cervantes-Sanchez","year":"2019","unstructured":"Cervantes-Sanchez F, Cruz-Aceves I, Hernandez-Aguirre A, Hernandez-Gonzalez MA, Solorio-Meza SE. Automatic segmentation of coronary arteries in x-ray angiograms using multiscale analysis and artificial neural networks. Appl Sci. 2019;9(24):5507.","journal-title":"Appl Sci"},{"key":"356_CR35","doi-asserted-by":"crossref","unstructured":"Wang W, Zhong J, Wu H, Wen Z, Qin J. Rvseg-net: An efficient feature pyramid cascade network for retinal vessel segmentation. In: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2020: 23rd International Conference, Lima, Peru, October 4\u20138, 2020, Proceedings, Part V 23, 2020;796\u2013805. Springer.","DOI":"10.1007\/978-3-030-59722-1_77"},{"issue":"2","key":"356_CR36","doi-asserted-by":"publisher","first-page":"1504","DOI":"10.1609\/aaai.v37i2.25236","volume":"37","author":"Z Li","year":"2023","unstructured":"Li Z, Li X, Yang L, Zhao B, Song R, Luo L, Li J, Yang J. Curriculum temperature for knowledge distillation. Proc AAAI Conf Artif Intell. 2023;37(2):1504\u201312. https:\/\/doi.org\/10.1609\/aaai.v37i2.25236.","journal-title":"Proc AAAI Conf Artif Intell"}],"container-title":["Health Information Science and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-025-00356-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13755-025-00356-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-025-00356-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T09:09:28Z","timestamp":1765444168000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13755-025-00356-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,19]]},"references-count":36,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["356"],"URL":"https:\/\/doi.org\/10.1007\/s13755-025-00356-4","relation":{},"ISSN":["2047-2501"],"issn-type":[{"value":"2047-2501","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,19]]},"assertion":[{"value":"22 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 June 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 June 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no Conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"39"}}