基于多尺度特征融合与自注意力的MRI脑胶质瘤分割算法
Glioma Segmentation Algorithm of MRI via Multi-Scale Feature Fusion and Self-Attention
摘要: 针对核磁共振图像(MRI)脑胶质瘤分割过程中存在特征信息丢失、边界模糊、错误分割等问题，设计一种脑胶质瘤分割算法。该算法基于DenseU-Net网络模型构建了混合型损失函数，提升网络的综合性能，并结合多尺度卷积并行合并结构，减少肿瘤边缘细节特征信息的丢失。借助部署在网络底层的池化单元，采集多感受野下脑胶质瘤的轮廓特征，提高特征信息的重复利用率，利用注意力机制使模型更加关注肿瘤区域。在脑胶质瘤数据集上进行本文算法有效性验证实验，本文算法的WT Dice值可以达到87.94%，证明了本文算法在脑胶质瘤分割任务中的有效性。
Abstract: To address issues such as feature information loss, blurred boundaries, and incorrect segmentation in the segmentation of gliomas from magnetic resonance imaging (MRI) scans, a glioma segmentation algorithm was designed. This method constructs a hybrid loss function based on the DenseU-Net network model to enhance the network’s overall performance. It also incorporates a multi-scale convolutional parallel merging structure to reduce the loss of detailed edge features in tumor margins. By utilizing pooling units deployed in the shallow layers of the network, it captures contour features of gliomas under multiple receptive fields, improving the reuse rate of feature information. The attention mechanism is employed to make the model focus more on tumor regions. Experiments on glioma datasets validate the effectiveness of this algorithm, achieving a WT Dice score of 87.94%, demonstrating the efficacy and superiority of the proposed network in glioma segmentation tasks.
文章引用：谭淇方. 基于多尺度特征融合与自注意力的MRI脑胶质瘤分割算法[J]. 图像与信号处理, 2026, 15(4): 532-543. https://doi.org/10.12677/jisp.2026.154047

参考文献

[1] 张奕涵, 柏正尧, 尤逸琳, 等. 自适应模态融合双编码器MRI脑肿瘤分割网络[J]. 中国图象图形学报, 2024, 29(3): 768-781.
[2] 孙家阔, 张荣, 郭立君, 汪建华. 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.
[3] 赵奕名, 李锵, 关欣. 组卷积轻量级脑肿瘤分割网络[J]. 中国图象图形学报, 2020, 25(10): 2159-2170.
[4] 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.
[5] Ghazouani, F., Vera, P. and Ruan, S. (2024) Efficient Brain Tumor Segmentation Using Swin Transformer and Enhanced Local Self-Attention. International Journal of Computer Assisted Radiology and Surgery, 19, 273-281.
https://doi.org/10.1007/s11548-023-03024-8
[6] Zhu, Z., He, X., Qi, G., Li, Y., Cong, B. and Liu, Y. (2023) Brain Tumor Segmentation Based on the Fusion of Deep Semantics and Edge Information in Multimodal MRI. Information Fusion, 91, 376-387.
https://doi.org/10.1016/j.inffus.2022.10.022
[7] Shelhamer, E., Long, J. and Darrell, T. (2017) Fully Convolutional Networks for Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 640-651.
https://doi.org/10.1109/tpami.2016.2572683
[8] Ronneberger, O., Fischer, P. and Brox, T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab, N., Hornegger, J., Wells, W. and Frangi, A., Eds., Lecture Notes in Computer Science, Springer International Publishing, 234-241.
https://doi.org/10.1007/978-3-319-24574-4_28
[9] Kermi, A., Mahmoudi, I. and Khadir, M.T. (2019) Deep Convolutional Neural Networks Using U-Net for Automatic Brain Tumor Segmentation in Multimodal MRI Volumes. In: Crimi, A., Bakas, S., Kuijf, H., Keyvan, F., Reyes, M. and van Walsum, T., Eds., Lecture Notes in Computer Science, Springer International Publishing, 37-48.
https://doi.org/10.1007/978-3-030-11726-9_4
[10] Zhou, Z., He, Z. and Jia, Y. (2020) AFPNet: A 3D Fully Convolutional Neural Network with Atrous-Convolution Feature Pyramid for Brain Tumor Segmentation via MRI Images. Neurocomputing, 402, 235-244.
https://doi.org/10.1016/j.neucom.2020.03.097
[11] Isensee, F., Jäger, P.F., Full, P.M., Vollmuth, P. and Maier-Hein, K.H. (2021) NNU-Net for Brain Tumor Segmentation. In: Crimi, A. and Bakas, S. Eds., Lecture Notes in Computer Science, Springer International Publishing, 118-132.
https://doi.org/10.1007/978-3-030-72087-2_11
[12] 柴文光, 李文浩, 闫敬文. 基于多任务的脑肿瘤MRI分割算法[J]. 光电子·激光, 2024, 35(9): 1001-1008.
[13] Singh, G. and Phophalia, A. (2022) Multimodal Brain Tumor Segmentation Using Modified UNet Architecture. In: Crimi, A. and Bakas, S., Eds., Lecture Notes in Computer Science, Springer International Publishing, 295-305.
https://doi.org/10.1007/978-3-031-08999-2_24
[14] Zhao, C., Liu, K., Chen, W., Pei, Z. and Feng, Y. (2022) Multi-Modality Brain Tumor Segmentation Network Based on Collaborative Feature Fusion. 2022 IEEE 17th Conference on Industrial Electronics and Applications (ICIEA), Chengdu, 16-19 December 2022, 1122-1127.
https://doi.org/10.1109/iciea54703.2022.10005998
[15] 毋小省, 杨奇鸿, 唐朝生, 等. 融合注意力机制的多模态脑肿瘤MR图像分割[J]. 计算机辅助设计与图形学学报, 2023, 35(9): 1429-1438.
[16] Shaikh, M., Anand, G., Acharya, G., Amrutkar, A., Alex, V. and Krishnamurthi, G. (2018) Brain Tumor Segmentation Using Dense Fully Convolutional Neural Network. In: Crimi, A., Bakas, S., Kuijf, H., Menze, B. and Reyes, M. Eds., Lecture Notes in Computer Science, Springer International Publishing, 309-319.
https://doi.org/10.1007/978-3-319-75238-9_27
[17] Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N. and Liang, J. (2019) UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation. IEEE Transactions on Medical Imaging, 39, 1856-1867.
https://doi.org/10.1109/tmi.2019.2959609
[18] 周涛, 侯森宝, 陆惠玲, 等. 探析U-Net的改进机制及其在医学图像分割的应用[J]. 生物医学工程学杂志, 2022, 39(4): 806-825.
[19] Senapati, P., Basu, A., Deb, M. and Dhal, K.G. (2024) Sharp Dense U-Net: An Enhanced Dense U-Net Architecture for Nucleus Segmentation. International Journal of Machine Learning and Cybernetics, 15, 2079-2094.
https://doi.org/10.1007/s13042-023-02017-y
[20] 万雨龙, 周冬明, 王长城, 等. 基于像素编码和空间注意力的多尺度医学图像分割方法[J]. 生物医学工程学杂志, 2024, 41(3): 511-519.
[21] Li, X., Chen, H., Qi, X., Dou, Q., Fu, C. and Heng, P. (2018) H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes. IEEE Transactions on Medical Imaging, 37, 2663-2674.
https://doi.org/10.1109/tmi.2018.2845918
[22] Chen, L., Papandreou, G., Kokkinos, I., Murphy, K. and Yuille, A.L. (2017) Deeplab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40, 834-848.
https://doi.org/10.1109/tpami.2017.2699184
[23] Wang, X., Girshick, R., Gupta, A. and He, K. (2018) Non-Local Neural Networks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-23 June 2018, 7794-7803.
https://doi.org/10.1109/cvpr.2018.00813
[24] 张耀, 刘艳君, 刘磊. 基于Inception模块的CNN-BiLSTM房颤检测与心拍分类算法[J]. 中国生物医学工程学报, 2024, 43(4): 447-454.
[25] 熊亮, 秦小林, 刘欣. 基于非局域注意力和多任务学习的胸部X线片肺实质分割方法[J]. 生物医学工程学杂志, 2023, 40(5): 912-919.
[26] He, K., Zhang, X., Ren, S. and Sun, J. (2015) Delving Deep into Rectifiers: Surpassing Human-Level Performance on Imagenet Classification. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 7-13 December 2015, 1026-1034.
https://doi.org/10.1109/iccv.2015.123
[27] Li, X., Sun, X., Meng, Y., Liang, J., Wu, F. and Li, J. (2020) Dice Loss for Data-Imbalanced NLP Tasks. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online, 5-10 July 2020, 465-476.
https://doi.org/10.18653/v1/2020.acl-main.45
[28] Wu, X., Huang, F. and Huang, H. (2022) Fast Stochastic Recursive Momentum Methods for Imbalanced Data Mining. 2022 IEEE International Conference on Data Mining (ICDM), Orlando, 28 November 2022-1 December 2022, 578-587.
https://doi.org/10.1109/icdm54844.2022.00068
[29] Isensee, F., Jaeger, P.F., Kohl, S.A.A., Petersen, J. and Maier-Hein, K.H. (2020) NNU-Net: A Self-Configuring Method for Deep Learning-Based Biomedical Image Segmentation. Nature Methods, 18, 203-211.
https://doi.org/10.1038/s41592-020-01008-z
[30] Wang, W., Chen, C., Ding, M., Yu, H., Zha, S. and Li, J. (2021) TransBTS: Multimodal Brain Tumor Segmentation Using Transformer. In: de Bruijne, M., et al. Eds., Lecture Notes in Computer Science, Springer International Publishing, 109-119.
https://doi.org/10.1007/978-3-030-87193-2_11
[31] Xie, E., Wang, W., Yu, Z., et al. (2021) SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers. Advances in Neural Information Processing Systems, 34, 12077-12090.
[32] Rehman, A., Usman, M., Shahid, A., et al. (2023) Selective Deeply Supervised Multi-Scale Attention Network for Brain Tumor Segmentation. Sensors, 23, 2346-2361.
https://doi.org/10.3390/s23042346