{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T23:16:59Z","timestamp":1789168619900,"version":"build-2803163510"},"reference-count":39,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Accurate polyp segmentation in colonoscopy supports early detection of colorectal cancer, but compact models under a student-only inference budget tend to lose boundary fidelity. BDKD-Net is a 3.72M-parameter compact student trained under a composite knowledge-distillation loss that combines a response signal, a boundary-probability signal restricted to a static teacher-derived boundary band, and an auxiliary detail-feature alignment; the teacher is used only during training and discarded at inference. In a uniform five-seed re-run on a locked Kvasir-SEG and CVC-ClinicDB development split, the same-architecture scratch student reaches Dev Dice 0.9210 \u00b1 0.0029 and boundary F1 at 3-pixel tolerance 0.7726 \u00b1 0.0057, while full BDKD-Net reaches 0.9300 \u00b1 0.0029 and 0.8008 \u00b1 0.0094. Boundary-probability distillation is the load-bearing signal: it has the highest mean Dev BF1t3 among the single KD signals and carries the boundary gain at no inference cost, with the full model reaching four-external Dice 0.8225 \u00b1 0.0071 at 1.76 GFLOPs and 140.9 FPS. Among the directly reproduced baselines, the higher-Dice Polyp-PVT (0.8417 \u00b1 0.0091) needs 6.8\u00d7 the parameters and 5.7\u00d7 the compute at roughly half the frame rate. BDKD-Net thus delivers a compact, boundary-faithful student that keeps most of the accuracy of much larger models at a fraction of their inference cost.<\/jats:p>","DOI":"10.3390\/jimaging12070306","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T16:16:18Z","timestamp":1783527378000},"page":"306","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["BDKD-Net: Boundary-Probability Knowledge Distillation for Compact Polyp Segmentation"],"prefix":"10.3390","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-9697-8550","authenticated-orcid":false,"given":"Tian","family":"Xia","sequence":"first","affiliation":[{"name":"College of Medical Instruments, Shanghai University of Medicine & Health Sciences, Shanghai 201318, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-9660-7218","authenticated-orcid":false,"given":"Jianhua","family":"Li","sequence":"additional","affiliation":[{"name":"College of Medical Instruments, Shanghai University of Medicine & Health Sciences, Shanghai 201318, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1642-0168","authenticated-orcid":false,"given":"Liping","family":"Sun","sequence":"additional","affiliation":[{"name":"College of Medical Instruments, Shanghai University of Medicine & Health Sciences, Shanghai 201318, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,8]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"687","DOI":"10.1056\/NEJMoa1100370","article-title":"Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths","volume":"366","author":"Zauber","year":"2012","journal-title":"N. Engl. J. Med."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1007\/978-3-030-37734-2_37","article-title":"Kvasir-SEG: A Segmented Polyp Dataset","volume":"Volume 11962","author":"Ro","year":"2020","journal-title":"MultiMedia Modeling"},{"key":"ref_3","first-page":"4037190","article-title":"A benchmark for endoluminal scene segmentation of colonoscopy images","volume":"2017","author":"Bernal","year":"2017","journal-title":"J. Healthc. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"2610","DOI":"10.1109\/TCYB.2022.3162873","article-title":"PolypSeg +: A Lightweight Context-Aware Network for Real-Time Polyp Segmentation","volume":"53","author":"Wu","year":"2023","journal-title":"IEEE Trans. Cybern."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Li, C., Xu, H., Zhu, X., Chen, H., Liu, X., Liu, Y., Tang, C., Chen, Z., Li, M., and Lu, S. (2026). HCA-Net: Hierarchical Contextual Attention Network for Lightweight and Accurate Polyp Segmentation. IEEE J. Biomed. Health Inform., Epub ahead of printing.","DOI":"10.1109\/JBHI.2026.3657790"},{"key":"ref_6","first-page":"263","article-title":"PraNet: Parallel Reverse Attention Network for Polyp Segmentation","volume":"Volume 12266","author":"Martel","year":"2020","journal-title":"Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2020"},{"key":"ref_7","first-page":"699","article-title":"Shallow Attention Network for Polyp Segmentation","volume":"Volume 12902","author":"Cattin","year":"2021","journal-title":"Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2021"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"9150015","DOI":"10.26599\/AIR.2023.9150015","article-title":"Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers","volume":"2","author":"Dong","year":"2023","journal-title":"CAAI Artif. Intell. Res."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"109555","DOI":"10.1016\/j.patcog.2023.109555","article-title":"Cross-level Feature Aggregation Network for Polyp Segmentation","volume":"140","author":"Zhou","year":"2023","journal-title":"Pattern Recognit."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Bui, N.T., Hoang, D.H., Nguyen, Q.T., Tran, M.T., and Le, N. (2024, January 3\u20138). MEGANet: Multi-Scale Edge-Guided Attention Network for Weak Boundary Polyp Segmentation. Proceedings of the 2024 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA.","DOI":"10.1109\/WACV57701.2024.00780"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"407","DOI":"10.1002\/mp.16584","article-title":"MGF-net: Multi-channel group fusion enhancing boundary attention for polyp segmentation","volume":"51","author":"Huang","year":"2024","journal-title":"Med. Phys."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"108770","DOI":"10.1016\/j.compbiomed.2024.108770","article-title":"C2E-Net: Cascade attention and context-aware cross-level fusion network via edge learning guidance for polyp segmentation","volume":"185","author":"Mao","year":"2025","journal-title":"Comput. Biol. Med."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"175041","DOI":"10.1088\/1361-6560\/acf090","article-title":"MCSF-Net: A multi-scale channel spatial fusion network for real-time polyp segmentation","volume":"68","author":"Liu","year":"2023","journal-title":"Phys. Med. Biol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge Distillation: A Survey","volume":"129","author":"Gou","year":"2021","journal-title":"Int. J. Comput. Vis."},{"key":"ref_15","unstructured":"Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the Knowledge in a Neural Network. arXiv."},{"key":"ref_16","unstructured":"Romero, A., Ballas, N., Kahou, S.E., Chassang, A., Gatta, C., and Bengio, Y. (2015, January 7\u20139). FitNets: Hints for Thin Deep Nets. Proceedings of the International Conference on Learning Representations (ICLR), San Diego, CA, USA."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Trinh, Q.H., Nguyen, M.V., and Thi, P.T.V. (2024, January 15\u201319). KDAS: Knowledge Distillation via Attention Supervision Framework for Polyp Segmentation. Proceedings of the 2024 IEEE International Conference on Multimedia and Expo (ICME), Niagara Falls, ON, Canada.","DOI":"10.1109\/ICME57554.2024.10687662"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e70050","DOI":"10.1049\/htl2.70050","article-title":"PRISM: Past-Regularized Iterative Self-Distillation with Momentum for Polyp Segmentation","volume":"12","author":"Erol","year":"2025","journal-title":"Healthc. Technol. Lett."},{"key":"ref_19","unstructured":"Wang, Z., Zhang, Y., Esrafilzadeh, D., Jalili, A.R., and Xiang, S. (2025). MicroAUNet: Boundary-Enhanced Multi-scale Fusion with Knowledge Distillation for Colonoscopy Polyp Image Segmentation. arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Qiao, J., and Wang, L. (2025, January 13\u201315). Edge-Guided and Feature-Aligned Coordinated Knowledge Distillation for Polyp Segmentation. Proceedings of the 2025 5th International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA), Beijing, China.","DOI":"10.1109\/CAIBDA65784.2025.11183003"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Rahman, M.M., Munir, M., and Marculescu, R. (2024, January 16\u201322). EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation. Proceedings of the 2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.01118"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"109777","DOI":"10.1016\/j.bspc.2026.109777","article-title":"LGPS: A lightweight GAN-based approach for polyp segmentation in colonoscopy images","volume":"118","author":"Tesema","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_23","unstructured":"Wu, Z., Ou, W., Zhang, L., Tan, P.S., Wu, D., Zhao, J., Fang, W., and Phan, R.C.W. (2026). DepthPolyp: Pseudo-Depth Guided Lightweight Segmentation for Real-Time Colonoscopy. arXiv."},{"key":"ref_24","unstructured":"Agnihotri, S., Majhi, S., and Nayak, D.R. (2026). Sharpening Lightweight Models for Generalized Polyp Segmentation: A Boundary Guided Distillation from Foundation Models. arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"3782","DOI":"10.1109\/TCSVT.2025.3621309","article-title":"De-LightSAM: Modality-Decoupled Lightweight SAM for Generalizable Medical Segmentation","volume":"36","author":"Xu","year":"2026","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_26","unstructured":"Tan, Z.Y., and Qasem, A. (2025). MEGANet-W: A Wavelet-Driven Edge-Guided Attention Framework for Weak Boundary Polyp Detection. arXiv."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"415","DOI":"10.1007\/s41095-022-0274-8","article-title":"PVT v2: Improved Baselines with Pyramid Vision Transformer","volume":"8","author":"Wang","year":"2022","journal-title":"Comput. Vis. Media"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1016\/j.compmedimag.2015.02.007","article-title":"WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians","volume":"43","author":"Bernal","year":"2015","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"3166","DOI":"10.1016\/j.patcog.2012.03.002","article-title":"Towards Automatic Polyp Detection with a Polyp Appearance Model","volume":"45","author":"Bernal","year":"2012","journal-title":"Pattern Recognit."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1007\/s11548-013-0926-3","article-title":"Toward embedded detection of polyps in WCE images for early diagnosis of colorectal cancer","volume":"9","author":"Silva","year":"2014","journal-title":"Int. J. Comput. Assist. Radiol. Surg."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Liu, G., Reda, F.A., Shih, K.J., Wang, T.C., Tao, A., and Catanzaro, B. (2018, January 8\u201314). Image Inpainting for Irregular Holes Using Partial Convolutions. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01252-6_6"},{"key":"ref_32","unstructured":"Loshchilov, I., and Hutter, F. (2019, January 6\u20139). Decoupled Weight Decay Regularization. Proceedings of the International Conference on Learning Representations, New Orleans, LA, USA."},{"key":"ref_33","unstructured":"Loshchilov, I., and Hutter, F. (2017, January 24\u201326). SGDR: Stochastic Gradient Descent with Warm Restarts. Proceedings of the International Conference on Learning Representations, Toulon, France."},{"key":"ref_34","unstructured":"Zhu, L. (2019). THOP: PyTorch-OpCounter, GitHub. Available online: https:\/\/github.com\/Lyken17\/pytorch-OpCounter."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.C. (2018, January 18\u201322). MobileNetV2: Inverted Residuals and Linear Bottlenecks. Proceedings of the 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT.","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Howard, A., Sandler, M., Chen, B., Wang, W., Chen, L.C., Tan, M., Chu, G., Vasudevan, V., Zhu, Y., and Pang, R. (November, January 27). Searching for MobileNetV3. Proceedings of the 2019 IEEE\/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref_37","unstructured":"Chen, L.C., Papandreou, G., Schroff, F., and Adam, H. (2017). Rethinking Atrous Convolution for Semantic Image Segmentation. arXiv."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Zhao, X., Jia, H., Pang, Y., Lv, L., Tian, F., Zhang, L., Sun, W., and Lu, H. (2023). M2SNet: Multi-scale in Multi-scale Subtraction Network for Medical Image Segmentation. arXiv.","DOI":"10.1016\/j.bspc.2023.105330"},{"key":"ref_39","first-page":"110","article-title":"SSFormer: Stepwise Feature Fusion: Local Guides Global","volume":"Volume 13433","author":"Wang","year":"2022","journal-title":"Proceedings of the Medical Image Computing and Computer Assisted Intervention\u2013MICCAI, Cham, Switzerland, 18\u201322 September 2022"}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/306\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T16:39:28Z","timestamp":1783528768000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/306"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,8]]},"references-count":39,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["jimaging12070306"],"URL":"https:\/\/doi.org\/10.3390\/jimaging12070306","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,8]]}}}