融合视觉状态空间与深度可分离卷积的脑胶质瘤MRI分割方法

Brain glioma MRI segmentation method integrating visual state space and depthwise separable convolution

摘要

目的：为提高对脑胶质瘤MRI图像的分割精度，增强对肿瘤全局浸润性及复杂边界的处理能力，提出一种状态空间增强的上下文与边界感知分割网络（SCB-UNet）。方法：首先，在编码器中引入VMamba模块，利用视觉状态空间机制高效建立图像全局依赖关系，以增强模型对肿瘤整体结构的感知能力。其次，在解码器中设计多尺度视觉状态空间前馈网络，通过融合全局上下文与局部多尺度特征，提升模型对肿瘤内部异质成分的区分能力。最后，构建深度可分离卷积扩展模块，在特征上采样前聚合局部上下文信息，以改善复杂边界和小病灶的细节恢复精度。结果：在BraTS2019和BraTS2023公开数据集上的实验结果表明，SCB-UNet在不同肿瘤区域上的平均Dice相似系数（DSC）分别达到85.94%和83.33%，平均95% Hausdorff（HD95）距离分别为4.18和5.13 mm。在BraTS2023数据集的消融实验中，相比U-Net基线模型，SCB-UNet模型的平均HD95降低1.46 mm。消融实验进一步验证各模块的有效性。结论：SCB-UNet模型为脑胶质瘤MRI图像的精准分割提供一种可行的技术途径，具有一定的临床应用潜力。

Abstract

Objective To propose a state space-enhanced context and boundary-aware segmentation network (SCB-UNet) for enhancing the segmentation accuracy of brain glioma MRI images and improving the model's ability to cope with the global infiltrative nature and complex boundaries of tumors. Methods A VMamba module was introduced into the encoder to model global image dependencies via a visual state space mechanism, thereby strengthening the model's perception of the overall tumor structure. Furthermore, a multi-scale visual state space feedforward network was established in the decoder to integrate global context with local multi-scale features, thus boosting the model's ability to distinguish heterogeneous components inside the tumor. Finally, a depthwise separable convolution expansion module is constructed to aggregate local contextual information before feature upsampling, which improves the accuracy of detail recovery for complex boundaries and small lesions. Results Experiments on the public BraTS2019 and BraTS2023 datasets show that the average Dice similarity coefficients of the SCB-UNet across different tumor regions reached 85.94% and 83.33%, respectively, and the average 95% Hausdorff distances were 4.18 and 5.13 mm, respectively. In the ablation experiment on the BraTS2023 dataset, compared with U-Net baseline model, the SCB-UNet reduced the average 95% Hausdorff distance by 1.46 mm, further verifying the effectiveness of each module. Conclusion The SCB-UNet provides a feasible technical solution for the accurate segmentation of brain glioma MRI images and demonstrates promising potential for clinical application.

关键词

脑胶质瘤 / MRI图像分割 / VMamba / 视觉状态空间 / 深度可分离卷积

Key words

brain glioma / MRI image segmentation / VMamba / visual state space / depthwise separable convolution

参考文献

[1] Yuan FN, Zhang ZX, Fang ZJ . An effective CNN and transformer complementary network for medical image segmentation[J]. Pattern Recognit, 2023, 136: 109228.

[2] 蔡林沁, 易文渊, 黄宇婷, 等 . 面向脑胶质瘤影像分析的混合现实技术[J]. 软件学报, 2022, 33(9): 3347-3369.

[3] Cai LQ, Yi WY, Huang YT, et al. Mixed reality technology for medical imaging analysis of glioma[J]. Journal of Software, 2022, 33(9): 3347-3369.

[4] 沈维高, 刘雨齐, 张骏, 等 . 白细胞介素33对恶性脑胶质瘤发生发展影响的生物信息学分析及其实验验证[J]. 吉林大学学报(医学版), 2025, 51(5): 1318-1332.

[5] Shen WG, Liu YQ, Zhang J, et al. Bioinformatics analysis on effect of interleukin—33 on occurrence and development of malignant brain glioma and its experimental validation[J]. Journal of Jilin University (Medicine Edition), 2025, 51(5): 1318-1332.

[6] Zhou HY, Guo JS, Zhang YH, et al. nnFormer: volumetric medical image segmentation via a 3D transformer [J]. IEEE Trans Image Process, 2023, 32: 4036-4045.

[7] Zhao T, Gu Y, Yang JW, et al. A foundation model for joint segmentation, detection and recognition of biomedical objects across nine modalities[J]. Nat Methods, 2025, 22(1): 166-176.

[8] Xiao HG, Li L, Liu QY, et al. Transformers in medical image segmentation: a review[J]. Biomed Signal Process Control, 2023, 84: 104791.

[9] 唐钰清, 庞彩凤, 周灵, 等 . 基于磁共振成像的影像组学在食管癌中的应用进展[J]. 磁共振成像, 2025, 16(9): 197-202.

[10] Tang YQ, Pang CF, Zhou L, et al. Progress in the application of MRI—based radiomics in esophageal cancer[J]. Chinese Journal of Magnetic Resonance Imaging, 2025, 16(9): 197-202.

[11] Azad R, Aghdam EK, Rauland A, et al. Medical image segmentation review: the success of U—Net[J]. IEEE Trans Pattern Anal Mach Intell, 2024, 46(12): 10076-10095.

[12] 王晓慧, 夏云飞 . 高级别脑胶质瘤放疗的现状与进展[J]. 广东医学, 2017, 38(1): 9-12.

[13] Wang XH, Xia YF . Current status and progress of radiotherapy for high—grade glioma[J]. Guangdong Medical Journal, 2017, 38(1): 9-12.

[14] 王玉丽, 赵子健 . 基于深度学习的脑图像分割算法研究综述[J]. 生物医学工程学杂志, 2020, 37(4): 721-729.

[15] Wang YL, Zhao ZJ . Research on brain image segmentation based on deep learning[J]. Journal of Biomedical Engineering, 2020, 37(4): 721-729.

[16] 江宗康, 吕晓钢, 张建新, 等 . MRI脑肿瘤图像分割的深度学习方法综述[J]. 中国图象图形学报, 2020, 25(2): 215-228.

[17] Jiang ZK, Lü XG, Zhang JX, et al. Review of deep learning methods for MRI brain tumor image segmentation[J]. Journal of Image and Graphics, 2020, 25(2): 215-228.

[18] 孙家阔, 张荣, 郭立君, 等 . 多尺度特征融合与加性注意力指导脑肿瘤MR图像分割[J]. 中国图象图形学报, 2023, 28(4): 1157-1172.

[19] Sun JK, Zhang R, Guo LJ, et al. Multi—scale feature fusion and additive attention guide brain tumor MR image segmentation[J]. Journal of Image and Graphics, 2023, 28(4): 1157-1172.

[20] 夏峰, 邵海见, 邓星 . 融合跨阶段深度学习的脑肿瘤MRI图像分割[J]. 中国图象图形学报, 2022, 27(3): 873-884.

[21] Xia F, Shao HJ, Deng X . Cross—stage deep—learning—based MRI fused images of human brain tumor segmentation[J]. Journal of Image and Graphics, 2022, 27(3): 873-884.

[22] Pei LM, Vidyaratne L, Rahman MM, et al. Context aware deep learning for brain tumor segmentation, subtype classification, and survival prediction using radiology images[J]. Sci Rep, 2020, 10(1): 19726.

[23] Wang WX, Chen C, Ding M, et al. TransBTS: multimodal brain tumor segmentation using transformer[C]// Medical Image Computing and Computer Assisted Intervention—MICCAI 2021. Cham: Springer International Publishing, 2021: 109-119.

[24] Liu X, Zhang CX, Huang FX, et al. Vision mamba: a comprehensive survey and taxonomy[J]. IEEE Trans Neural Netw Learn Syst, 2026, 37(2): 505-525.

[25] Wang ZH, Kong FH, Feng S, et al. Is mamba effective for time series forecasting?[J]. Neurocomputing, 2025, 619: 129178.

[26] Bakas S, Reyes M, Jakab A, et al. Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge[EB/OL]. (2019—04—23). https://arxiv.org/abs/1811.02629.

[27] Li HW, Conte GM, Hu QQ, et al. The brain tumor segmentation (BraTS) challenge 2023: brain MR image synthesis for tumor segmentation (BraSyn)[EB/OL]. (2024—11—24). https://arxiv.org/abs/2305.09011.

[28] Zhang Y, Lu Y, Chen WK, et al. MSMANet: a multi—scale mesh aggregation network for brain tumor segmentation[J]. Appl Soft Comput, 2021, 110: 107733.

[29] Liu ZH, Tong L, Chen L, et al. CANet: context aware network for brain glioma segmentation[J]. IEEE Trans Med Imaging, 2021, 40(7): 1763-1777.

[30] Rosas—Gonzalez S, Birgui—Sekou T, Hidane M, et al. Asymmetric ensemble of asymmetric U—Net models for brain tumor segmentation with uncertainty estimation[J]. Front Neurol, 2021, 12: 609646.

[31] Chen C, Liu XP, Ding M, et al. 3D dilated multi—fiber network for real—time brain tumor segmentation in MRI[C]// Medical Image Computing and Computer Assisted Intervention—MICCAI 2019. Cham: Springer International Publishing, 2019: 184-192.

[32] Luo ZR, Jia ZD, Yuan ZM, et al. HDC—Net: hierarchical decoupled convolution network for brain tumor segmentation[J]. IEEE J Biomed Health Inform, 2021, 25(3): 737-745.

[33] Valanarasu JM, Sindagi VA, Hacihaliloglu I, et al. KiU—Net: overcomplete convolutional architectures for biomedical image and volumetric segmentation[J]. IEEE Trans Med Imaging, 2022, 41(4): 965-976.

[34] 黄敬腾, 李锵, 关欣 . 一种用于脑肿瘤分割的改进U形网络[J]. 工程科学学报, 2023, 45(6): 1003-1012.

[35] Huang JT, Li Q, Guan X . An improved U—shaped network for brain tumor segmentation[J]. Chinese Journal of Engineering, 2023, 45(6): 1003-1012.

[36] Sheng N, Liu DW, Zhang JX, et al. Second—order ResU—Net for automatic MRI brain tumor segmentation[J]. Math Biosci Eng, 2021, 18(5): 4943-4960.

[37] Ruan JC, Li JC, Xiang SC . VM—UNet: vision mamba UNet for medical image segmentation[J/OL]. ACM Trans Multimed Comput Commun Appl. (2025—09—16). https://doi.org/10.1145/3767748.

[38] Ma J, Li FF, Wang B . U—mamba: enhancing long—range dependency for biomedical image segmentation[EB/OL]. (2024—01—09). https://arxiv.org/abs/2401.04722.

基金资助

国家自然科学基金(82173388)