BKMNet: A boundary-enhanced state space model with Kalman filtering for adenomatous polyp segmentation

Jiaoju Wang; Bing Tan; Jingming Li; Yiming Shu; Meiling Tang; Binghua Tao; Muzhou Hou; Shuijiao Chen; Jiaoju Wang; Bing Tan; Jingming Li; Yiming Shu; Meiling Tang; Binghua Tao; Muzhou Hou; Shuijiao Chen

doi:10.3934/era.2026291

Electronic Research Archive

2026, Volume 34, Issue 9: 6658-6687. doi: 10.3934/era.2026291

Previous Article Next Article

Research article Special Issues

BKMNet: A boundary-enhanced state space model with Kalman filtering for adenomatous polyp segmentation

1.
School of Mathematics and Statistics, Nanyang Normal University, Nanyang 473061, China
2.
Applied Mathematics Center of Henan Province, Nanyang 473061, China
3.
School of Civil Engineering and Architecture, Nanyang Normal University, Nanyang 473061, China
4.
School of Mathematics and Statistics, Central South University, Changsha 410083, China
5.
School of Computer Science and Engineering, Hunan University of Information Technology, Changsha 410151, China
6.
Department of Gastroenterology, Xiangya Hospital of Central South University, Changsha 410008, China

Received: 27 May 2026 Revised: 12 July 2026 Accepted: 16 July 2026 Published: 04 August 2026

Colorectal cancer is one of the leading causes of cancer-related mortality, and both early detection and the removal of adenomatous polyps are effective treatment measures. However, identifying adenomatous polyps remains challenging due to the diverse polyps, boundaries, and intestinal mucosal background. Therefore, we have proposed a boundary-enhanced adenomatous polyp segmentation network with Kalman-mamba (BKMNet), which integrates visual state space mechanisms and Kalman filtering for precise segmentation and identification of adenomatous polyps. Specifically, we adopted a multi-frequency combined attention block to enhance feature representation by focusing on varying channel frequency components. A fast visual state space module extracts multi-level feature maps, the partial channel transformation aggregates the surrounding contextual information, and the structure-aware local-enhanced residual state space module enhances structure awareness. A Kalman-mamba layer aims for long-range dependency modeling. The decoder has a dual attention module with multi-level features, and a refined differential and boundary-enhanced module performs edge detection and global edge feature learning. Results demonstrated that BKMNet outperforms state-of-the-art methods in terms of segmentation accuracy and clinical applicability. This study provides a reliable auxiliary diagnostic tool for supporting clinicians in improving diagnostic efficiency and reducing misdiagnosis rates.
- mamba,
- Kalman filtering,
- boundary learning,
- segmentation,
- adenomatous polyps
Citation: Jiaoju Wang, Bing Tan, Jingming Li, Yiming Shu, Meiling Tang, Binghua Tao, Muzhou Hou, Shuijiao Chen. BKMNet: A boundary-enhanced state space model with Kalman filtering for adenomatous polyp segmentation[J]. Electronic Research Archive, 2026, 34(9): 6658-6687. doi: 10.3934/era.2026291

Related Papers:

Abstract

Colorectal cancer is one of the leading causes of cancer-related mortality, and both early detection and the removal of adenomatous polyps are effective treatment measures. However, identifying adenomatous polyps remains challenging due to the diverse polyps, boundaries, and intestinal mucosal background. Therefore, we have proposed a boundary-enhanced adenomatous polyp segmentation network with Kalman-mamba (BKMNet), which integrates visual state space mechanisms and Kalman filtering for precise segmentation and identification of adenomatous polyps. Specifically, we adopted a multi-frequency combined attention block to enhance feature representation by focusing on varying channel frequency components. A fast visual state space module extracts multi-level feature maps, the partial channel transformation aggregates the surrounding contextual information, and the structure-aware local-enhanced residual state space module enhances structure awareness. A Kalman-mamba layer aims for long-range dependency modeling. The decoder has a dual attention module with multi-level features, and a refined differential and boundary-enhanced module performs edge detection and global edge feature learning. Results demonstrated that BKMNet outperforms state-of-the-art methods in terms of segmentation accuracy and clinical applicability. This study provides a reliable auxiliary diagnostic tool for supporting clinicians in improving diagnostic efficiency and reducing misdiagnosis rates.

References

[1]	D. S. Chan, M. Cariolou, G. Markozannes, K. Balducci, R. Vieira, S. Kiss, et al., Post-diagnosis dietary factors, supplement use and colorectal cancer prognosis: A global cancer update programme (cup global) systematic literature review and meta-analysis, Int. J. Cancer, 155 (2024), 445–470. https://doi.org/10.1002/ijc.34906 doi: 10.1002/ijc.34906
[2]	J. Bond, Clinical relevance of the small colorectal polyp, Endoscopy, 33 (2001), 454–457. https://doi.org/10.1055/s-2001-14266 doi: 10.1055/s-2001-14266
[3]	L. Shi, H. Li, S. Li, S. Lin, Y. Wu, Pseudoinvasion and squamous metaplasia/morules in colorectal adenomatous polyp: A case report and literature review, Diagn. Pathol., 19 (2024), 126. https://doi.org/10.1186/s13000-024-01535-9 doi: 10.1186/s13000-024-01535-9
[4]	G. Yue, P. Wei, Y. Liu, Y. Luo, J. Du, T. Wang, Automated endoscopic image classification via deep neural network with class imbalance loss, IEEE Trans. Instrum. Meas., 72 (2023), 1–11. https://doi.org/10.1109/tim.2023.3264047 doi: 10.1109/tim.2023.3264047
[5]	L. Lin, G. Lv, B. Wang, C. Xu, J. Liu, Polyp-lvt: Polyp segmentation with lightweight vision transformers, Knowl. Based Syst., 300 (2024), 112181. https://doi.org/10.1016/j.knosys.2024.112181 doi: 10.1016/j.knosys.2024.112181
[6]	A. Leufkens, M. Van Oijen, F. Vleggaar, P. Siersema, Factors influencing the miss rate of polyps in a back-to-back colonoscopy study, Endoscopy, 44 (2012), 470–475. https://doi.org/10.1055/s-0031-1291666 doi: 10.1055/s-0031-1291666
[7]	Y. Teng, Y. Yang, J. Yang, Q. Lu, J. Shi, J. Xu, et al., Association between triglyceride-glucose index and colorectal polyps: A retrospective cross-sectional study, World J. Gastroenterol. Endosc., 16 (2024), 55. https://doi.org/10.4253/wjge.v16.i2.55 doi: 10.4253/wjge.v16.i2.55
[8]	D. Hong, B. Zhang, X. Li, Y. Li, C. Li, J. Yao, et al., Spectralgpt: Spectral remote sensing foundation model, IEEE Trans. Pattern Anal. Mach. Intell., 46 (2024), 5227–5244. https://doi.org/10.1109/tpami.2024.3362475 doi: 10.1109/tpami.2024.3362475
[9]	D. Hong, C. Li, N. Yokoya, B. Zhang, X. Jia, A. Plaza, et al., Hyperspectral imaging, Nat. Rev. Methods Primers, 6 (2026), 19. https://doi.org/10.1038/s43586-026-00470-x doi: 10.1038/s43586-026-00470-x
[10]	P. Singh, Y. Huang, Akdc: Ambiguous kernel distance clustering algorithm for COVID-19 ct scans analysis, IEEE Trans. Syst. Man Cybern. Syst., 54 (2024), 6218–6229. https://doi.org/10.1109/tsmc.2024.3418411 doi: 10.1109/tsmc.2024.3418411
[11]	P. Singh, Y. Huang, An ambiguous edge detection method for computed tomography scans of coronavirus disease 2019 cases, IEEE Trans. Syst. Man Cybern. Syst., 54 (2023), 352–364. https://doi.org/10.1109/tsmc.2023.3307393 doi: 10.1109/tsmc.2023.3307393
[12]	P. Singh, S. S. Bose, A quantum-clustering optimization method for COVID-19 ct scan image segmentation, Expert Syst. Appl., 185 (2021), 115637. https://doi.org/10.1016/j.eswa.2021.115637 doi: 10.1016/j.eswa.2021.115637
[13]	P. Singh, S. S. Bose, Ambiguous d-means fusion clustering algorithm based on ambiguous set theory: Special application in clustering of ct scan images of COVID-19, Knowl. Based Syst., 231 (2021), 107432. https://doi.org/10.1016/j.eswa.2021.115637 doi: 10.1016/j.eswa.2021.115637
[14]	Z. Zhu, H. Wang, G. Qi, Y. Li, N. Mazur, Y. Liu, et al., A survey on lightweight technology of neural networks for medical image segmentation, Pattern Recognit., 179 (2026), 113870. https://doi.org/10.1016/j.patcog.2026.113870 doi: 10.1016/j.patcog.2026.113870
[15]	Y. Ding, S. Li, H. Li, G. Qi, B. Cong, Y. Gong, et al., Physical regularization loss: Integrating physical knowledge to image segmentation, Int. J. Comput. Vis., 134 (2026), 137. https://doi.org/10.1007/s11263-026-02776-5 doi: 10.1007/s11263-026-02776-5
[16]	Z. Zhu, Z. Zhang, G. Qi, Y. Li, P. Yang, Y. Liu, Probability map-guided network for 3d volumetric medical image segmentation, IEEE Trans. Image Process., 34 (2025), 7222–7234. https://doi.org/10.1109/tip.2025.3623259 doi: 10.1109/tip.2025.3623259
[17]	X. Zhao, L. Zhang, H. Lu, Automatic polyp segmentation via multi-scale subtraction network, in Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Springer, (2021), 120–130. https://doi.org/10.1007/978-3-030-87193-2_12
[18]	W. Zhang, C. Fu, Y. Zheng, F. Zhang, Y. Zhao, C. M. Sham, Hsnet: A hybrid semantic network for polyp segmentation, Comput. Biol. Med., 150 (2022), 106173. https://doi.org/10.1016/j.compbiomed.2022.106173 doi: 10.1016/j.compbiomed.2022.106173
[19]	H. Wu, Z. Zhao, Z. Wang, Meta-unet: Multi-scale efficient transformer attention unet for fast and high-accuracy polyp segmentation, IEEE Trans. Autom. Sci. Eng. 21 (2023), 4117–4128. https://doi.org/10.1109/tase.2023.3292373 doi: 10.1109/tase.2023.3292373
[20]	M. Wang, X. An, Z. Pei, N. Li, L. Zhang, G. Liu, et al., An efficient multi-task synergetic network for polyp segmentation and classification, IEEE J. Biomed. Health Inform., 28 (2023), 1228–1239. https://doi.org/10.1109/jbhi.2023.3273728 doi: 10.1109/jbhi.2023.3273728
[21]	L. Meng, Y. Li, W. Duan, Three-stage polyp segmentation network based on reverse attention feature purification with pyramid vision transformer, Comput. Biol. Med., 179 (2024), 108930. https://doi.org/10.1016/j.compbiomed.2024.108930 doi: 10.1016/j.compbiomed.2024.108930
[22]	Z. Wu, H. Chen, X. Xiong, S. Wu, H. Li, X. Zhou, Bmanet: Boundary-guided multi-level attention network for polyp segmentation in colonoscopy images, Biomed. Signal Process. Control, 105 (2025), 107524. https://doi.org/10.1016/j.bspc.2025.107524 doi: 10.1016/j.bspc.2025.107524
[23]	S. Wang, S. Lin, F. Sun, X. Li, Multi-feature fusion for accurate polyp segmentation using pyramid visual transformers, Expert Syst. Appl., 280 (2025), 127558. https://doi.org/10.1016/j.eswa.2025.127558 doi: 10.1016/j.eswa.2025.127558
[24]	Y. Ma, Y. Liu, J. Cheng, H. Zhan, G. Qi, X. Zhang, et al., Cihm: Context-insight hybrid mamba for efficient medical image segmentation, Vis. Intell., 4 (2026), 12. https://doi.org/10.1007/s44267-026-00113-5 doi: 10.1007/s44267-026-00113-5
[25]	J. Ruan, J. Li, S. Xiang, Vm-unet: Vision mamba unet for medical image segmentation, preprint, arXiv: 2402.02491.
[26]	X. Zhu, W. Wang, C. Zhang, H. Wang, Polyp-mamba: A hybrid multi-frequency perception gated selection network for polyp segmentation, Inf. Fusion, 115 (2025), 102759. https://doi.org/10.1016/j.inffus.2024.102759 doi: 10.1016/j.inffus.2024.102759
[27]	O. Ronneberger, P. Fischer, T. Brox, U-net: Convolutional networks for biomedical image segmentation, in Medical Image Computing and Computer-assisted Intervention–MICCAI 2015: 18th International Conference, Springer, (2015), 234–241. https://doi.org/10.1007/978-3-319-24574-4_28
[28]	M. Shi, S. Lin, Q. Yi, J. Weng, A. Luo, Y. Zhou, Lightweight context-aware network using partial-channel transformation for real-time semantic segmentation, IEEE Trans. Intell. Transp. Syst., 25 (2024), 7401–7416. https://doi.org/10.1109/tits.2023.3348631 doi: 10.1109/tits.2023.3348631
[29]	D. Fan, G. Ji, T. Zhou, G. Chen, H. Fu, J. Shen, et al., Pranet: Parallel reverse attention network for polyp segmentation, in International Conference on Medical Image Computing and Computer-assisted Intervention, Springer, (2020), 263–273. https://doi.org/10.1007/978-3-030-59725-2_26
[30]	W. Zhang, B. Su, C. Huangfu, L. Zhang, A review of colorectal polyp segmentation methods, J. Comput. Electron. Inf. Manag., 20 (2026), 111–116. https://doi.org/10.54097/pqj8hf46 doi: 10.54097/pqj8hf46
[31]	H. He, X. Li, G. Cheng, J. Shi, Y. Tong, G. Meng, et al., Enhanced boundary learning for glass-like object segmentation, in Proceedings of the IEEE/CVF International Conference on Computer Vision, (2021), 15859–15868. https://doi.org/10.1109/iccv48922.2021.01556
[32]	D. Xie, Y. Zhang, X. Tian, L. Xu, L. Duan, L. Tian, Bgfe-net: A boundary-guided feature enhancement network for segmentation of targets with fuzzy boundaries, Neurocomputing, 618 (2025), 129127. https://doi.org/10.1016/j.neucom.2024.129127 doi: 10.1016/j.neucom.2024.129127
[33]	Y. Lin, D. Zhang, X. Fang, Y. Chen, K. Cheng, H. Chen, Rethinking boundary detection in deep learning models for medical image segmentation, in International Conference on Information Processing in Medical Imaging, Springer, (2023), 730–742. https://doi.org/10.1016/j.media.2025.103615
[34]	Z. Wu, X. Zhang, F. Li, S. Wang, L. Huang, J. Li, W-Net: A boundary-enhanced segmentation network for stroke lesions, Expert Syst. Appl., 230 (2023), 120637. https://doi.org/10.1016/j.eswa.2023.120637 doi: 10.1016/j.eswa.2023.120637
[35]	S. Li, X. Tang, B. Cao, Y. Peng, X. He, S. Ye, et al., Boundary guided network with two-stage transfer learning for gastrointestinal polyps segmentation, Expert Syst. Appl., 240 (2024), 122503. https://doi.org/10.1016/j.eswa.2023.122503 doi: 10.1016/j.eswa.2023.122503
[36]	G. Xu, J. Li, G. Gao, H. Lu, J. Yang, D. Yue, Lightweight real-time semantic segmentation network with efficient transformer and cnn, IEEE Trans. Intell. Transp. Syst., 24 (2023), 15897–15906. https://doi.org/10.1109/tits.2023.3248089 doi: 10.1109/tits.2023.3248089
[37]	B. Dong, W. Wang, D. Fan, J. Li, H. Fu, L. Shao, Polyp-pvt: Polyp segmentation with pyramid vision transformers, CAAI Artif. Intell. Res., 2 (2023), 9150015. https://doi.org/10.26599/air.2023.9150015 doi: 10.26599/air.2023.9150015
[38]	F. Tang, B. Nian, J. Ding, W. Ma, Q. Quan, C. Dong, et al., Mobile u-vit: Revisiting large kernel and u-shaped vit for efficient medical image segmentation, in Proceedings of the 33rd ACM International Conference on Multimedia, (2025), 3408–3417. https://doi.org/10.1145/3746027.3755076
[39]	H. Sun, Y. Zhang, L. Xu, S. Jin, Y. Chen, Ultra-high resolution segmentation via boundary-enhanced patch-merging transformer, in Proceedings of the AAAI Conference on Artificial Intelligence, (2025), 7087–7095. https://doi.org/10.1609/aaai.v39i7.32761
[40]	J. Chen, Y. Lu, Q. Yu, X. Luo, E. Adeli, Y. Wang, et al., Transunet: Transformers make strong encoders for medical image segmentation, preprint, arXiv: 2102.04306.
[41]	N. Zhang, L. Yu, D. Zhang, W. Wu, S. Tian, X. Kang, et al., Ct-net: Asymmetric compound branch transformer for medical image segmentation, Neural Netw., 170 (2024), 298–311. https://doi.org/10.2139/ssrn.4331189 doi: 10.2139/ssrn.4331189
[42]	W. Liao, Y. Zhu, X. Wang, C. Pan, Y. Wang, L. Ma, Lightm-unet: Mamba assists in lightweight unet for medical image segmentation, preprint, arXiv: 2403.05246.
[43]	V. T. Nguyen, V. T. Pham, T. T. Tran, Ac-mambaseg: An adaptive convolution and mamba-based architecture for enhanced skin lesion segmentation, in International Conference on Green Technology and Sustainable Development, Springer, (2024), 13–26. https://doi.org/10.1007/978-3-031-76197-3_2
[44]	Z. Zhang, Q. Ma, T. Zhang, J. Chen, H. Zheng, W. Gao, Switch-umamba: Dynamic scanning vision mamba unet for medical image segmentation, Med. Image Anal., 107 (2025), 103792. https://doi.org/10.1016/j.media.2025.103792 doi: 10.1016/j.media.2025.103792
[45]	Z. Wang, J. Zheng, Y. Zhang, G. Cui, L. Li, Mamba-unet: Unet-like pure visual mamba for medical image segmentation, preprint, arXiv: 2402.05079.
[46]	T. Hussain, H. Shouno, M. A. Mohammed, H. A. Marhoon, T. Alam, Dcssga-unet: Biomedical image segmentation with densenet channel spatial and semantic guidance attention, Knowl.-Based Syst., 314 (2025), 113233. https://doi.org/10.1016/j.knosys.2025.113233 doi: 10.1016/j.knosys.2025.113233
[47]	T. Zhou, Y. Zhou, K. He, C. Gong, J. Yang, H. Fu, et al., Cross-level feature aggregation network for polyp segmentation, Pattern Recognit., 140 (2023), 109555. https://doi.org/10.1016/j.patcog.2023.109555 doi: 10.1016/j.patcog.2023.109555
[48]	H. Huang, Z. Chen, Y. Zou, M. Lu, C. Chen, Y. Song, et al., Channel prior convolutional attention for medical image segmentation, Comput. Biol. Med., 178 (2024), 108784. https://doi.org/10.1016/j.compbiomed.2024.108784 doi: 10.1016/j.compbiomed.2024.108784
[49]	G. Liu, S. Yao, D. Liu, B. Chang, Z. Chen, J. Wang, et al., Cafe-net: Cross-attention and feature exploration network for polyp segmentation, Expert Syst. Appl., 238 (2024), 121754. https://doi.org/10.1016/j.eswa.2023.121754 doi: 10.1016/j.eswa.2023.121754
[50]	J. H. Nam, N. S. Syazwany, S. J. Kim, S. C. Lee, Modality-agnostic domain generalizable medical image segmentation by multi-frequency in multi-scale attention, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2024), 11480–11491. https://doi.org/10.1109/cvpr52733.2024.01091
[51]	D. Jha, P. H. Smedsrud, M. A. Riegler, P. Halvorsen, T. De Lange, D. Johansen, et al., Kvasir-seg: A segmented polyp dataset, in International Conference on Multimedia Modeling, Springer, (2019), 451–462. https://doi.org/10.1007/978-3-030-37734-2_37
[52]	J. Bernal, F. J. Sánchez, G. Fernández-Esparrach, D. Gil, C. Rodríguez, F. Vilariño, Wm-dova maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians, Comput. Med. Imaging Graph., 43 (2015), 99–111. https://doi.org/10.1016/j.compmedimag.2015.02.007 doi: 10.1016/j.compmedimag.2015.02.007
[53]	N. Tajbakhsh, S. R. Gurudu, J. Liang, Automated polyp detection in colonoscopy videos using shape and context information, IEEE Trans. Med. Imaging, 35 (2015), 630–644. https://doi.org/10.1109/tmi.2015.2487997 doi: 10.1109/tmi.2015.2487997

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)