|
|
基于扩散模型的皮肤病变分割方法
|
Abstract:
皮肤病变分割在临床诊断中扮演着重要角色。为了加强对多尺度上下文信息的学习，提高模型在皮肤病变边缘处的分割准确度，本研究提出一种新的基于扩散模型的皮肤病变分割框架DCA-SegDiff。该框架将双交叉注意模块集成到网络的跳跃连接部分，能够同时接收来自不同层编码器的多尺度特征信息，同时与解码器部分的挤压激励机制结合使用，以充分利用上下文特征信息。同时，采用基于Haar小波的下采样模块，以保留更多边缘细节信息。此外，在编码器深层采用高效增强多尺度注意力模块，通过跨空间学习的方法来融合多尺度特征信息。为验证模型的优越性，在ISIC2018、PH2、HAM10000三个皮肤病变分割数据集上进行实验，主要评价指标Dice分别达到了0.8944、0.9446、0.9445。结果表明，DCA-SegDiff模型在三个数据集上表现优于其他现有模型，而且相比基线模型的参数量显著减少，证明了其在皮肤病变分割任务中的有效性和泛化能力。
Segmentation of skin lesions plays an important role in clinical diagnosis. In this paper, a new skin lesion segmentation framework based on diffusion model, DCA-SegDiff, is proposed to enhance the learning of multi-scale context information to improve the segmentation accuracy of the model at the edge of skin lesions. The dual cross-attention module is integrated into the skip connection part of the denoising network in this framework, which can receive multi-scale feature information from different encoders at the same time. Meanwhile, the down-sampling module based on Haar wavelet is adopted to retain more edge detail information. In addition, an efficient enhanced multi-scale attention module is used in the deep layer of the encoder to integrate multi-scale feature information by cross-space learning method. In order to verify the superiority of the model, experiments are conducted on three skin lesion segmentation datasets, ISIC2018, PH2, and HAM10000, and Dice reaches 0.8944, 0.9446, and 0.9445, respectively. The results show that the DCA-SegDiff model outperforms other existing models on the three datasets, and the number of parameters is significantly reduced compared to the Baseline, demonstrating its effectiveness and generalization ability in the skin lesion segmentation task.
| [1] | Feigelson, H.S., Powers, J.D., Kumar, M., Carroll, N.M., Pathy, A. and Ritzwoller, D.P. (2019) Melanoma Incidence, Recurrence, and Mortality in an Integrated Healthcare System: A Retrospective Cohort Study. Cancer Medicine, 8, 4508-4516. https://doi.org/10.1002/cam4.2252 |
| [2] | 陈旋. 基于深度学习的皮肤病图像分割算法研究[D]: [硕士学位论文]. 贵阳: 贵州大学, 2024. |
| [3] | 胡越黎, 曹家麟. 皮肤显微图像的自适应分割[J]. 计算机工程, 2006, 32(10): 175-177, 182. |
| [4] | Yasmin, J. and Mohamed Sadiq, M. (2012) An Improved Iterative Segmentation Algorithm Using Canny Edge Detector with Iterative Median Filter for Skin Lesion Border Detection. International Journal of Computer Applications, 50, 37-42. https://doi.org/10.5120/7779-0865 |
| [5] | 徐国雄, 王海娜, 胡进贤, 等. 一种基于混合方法的彩色图像分割算法——结合分水岭算法、种子区域生长和区域合并的混合方法[J]. 计算机科学与应用, 2013, 3(2): 134-138. |
| [6] | Ronneberger, O., Fischer, P. and Brox, T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention, Berlin, 5-9 October 2015, 234-241. https://doi.org/10.1007/978-3-319-24574-4_28 |
| [7] | Diakogiannis, F.I., Waldner, F., Caccetta, P. and Wu, C. (2020) ResUNet-A: A Deep Learning Framework for Semantic Segmentation of Remotely Sensed Data. ISPRS Journal of Photogrammetry and Remote Sensing, 162, 94-114. https://doi.org/10.1016/j.isprsjprs.2020.01.013 |
| [8] | Oktay, O., Schlemper, J., Folgoc, L., et al. (2024) Attention U-Net: Learning Where to Look for the Pancreas. https://arxiv.org/abs/1804.03999 |
| [9] | Dosovitskiy, A., Beyer, L., Kolesnikov, A., et al. (2021) An Image Is Worth 16×16 Words: Transformers for Image Recognition at Scale. Proceedings of International Conference on Learning Representations, Virtual, 3-7 May 2021. https://openreview.net/forum?id=YicbFdNTTy |
| [10] | Chen, J., Lu, Y., Yu, Q., et al. (2024) TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation. https://arxiv.org/abs/2102.04306 |
| [11] | Valanarasu, J.M.J., Oza, P., Hacihaliloglu, I., et al. (2020) Medical Transformer: Gated Axial-Attention for Medical Image Segmentation. Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention, Strasbourg, 27 September-1 October 2020, 6840-6851. |
| [12] | 杨澜. 一种基于TransUNet的双分支并行医学图像分割模型[J]. 计算机科学与应用, 2024, 14(10): 74-84. |
| [13] | Ho, J., Jain, A. and Abbeel, P. (2022) Denoising Diffusion Probabilistic Models. Proceedings of Advances in Neural Information Processing Systems, Vancouver, 8-11 December 2022, 6840-6851. |
| [14] | Wolleb, J., Sandkühler, R., Bieder, F., et al. (2022) Diffusion Models for Implicit Image Segmentation Ensembles. Proceedings of International Conference on Medical Imaging with Deep Learning, Zurich, 6-8 July 2022, 1336-1348. |
| [15] | Wu, J.D., Fu, R., Fang, H.H., et al. (2024) MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic Model. Proceedings of International Conference on Medical Imaging with Deep Learning, Paris, 3-5 July 2024, 1623-1639. |
| [16] | Wu, J., Ji, W., Fu, H., Xu, M., Jin, Y. and Xu, Y. (2024) Medsegdiff-v2: Diffusion-Based Medical Image Segmentation with Transformer. Proceedings of the AAAI Conference on Artificial Intelligence, 38, 6030-6038. https://doi.org/10.1609/aaai.v38i6.28418 |
| [17] | Bozorgpour, A., Sadegheih, Y., Kazerouni, A., Azad, R. and Merhof, D. (2023) DermoSegDiff: A Boundary-Aware Segmentation Diffusion Model for Skin Lesion Delineation. Proceedings of International Workshop on Predictive Intelligence in Medicine, Vancouver, 8 October 2023, 146-158. https://doi.org/10.1007/978-3-031-46005-0_13 |
| [18] | Yu, F., Koltun, V. and Funkhouser, T. (2017) Dilated Residual Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 472-480. https://doi.org/10.1109/cvpr.2017.75 |
| [19] | Shen, Z., Zhang, M., Zhao, H., et al. (2021) Efficient Attention: Attention with Linear Complexities. Proceedings of Winter Conference on Applications of Computer Vision, 3-8 January 2021, 3531-3539. |
| [20] | Hu, J., Shen, L. and Sun, G. (2018) Squeeze-and-Excitation Networks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-22 June 2018, 7132-7141. https://doi.org/10.1109/cvpr.2018.00745 |
| [21] | Xu, G., Liao, W., Zhang, X., Li, C., He, X. and Wu, X. (2023) Haar Wavelet Downsampling: A Simple but Effective Downsampling Module for Semantic Segmentation. Pattern Recognition, 143, Article ID: 109819. https://doi.org/10.1016/j.patcog.2023.109819 |
| [22] | Ouyang, D., He, S., Zhang, G., Luo, M., Guo, H., Zhan, J., et al. (2023) Efficient Multi-Scale Attention Module with Cross-Spatial Learning. ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, 4-10 June 2023, 1-5. https://doi.org/10.1109/icassp49357.2023.10096516 |
| [23] | Ates, G.C., Mohan, P. and Celik, E. (2023) Dual Cross-Attention for Medical Image Segmentation. Engineering Applications of Artificial Intelligence, 126, Article ID: 107139. https://doi.org/10.1016/j.engappai.2023.107139 |
| [24] | Codella, N.C.F., Gutman, D., Celebi, M.E., Helba, B., Marchetti, M.A., Dusza, S.W., et al. (2018) Skin Lesion Analysis toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC). 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, 4-7 April 2018, 168-172. https://doi.org/10.1109/isbi.2018.8363547 |
| [25] | Mendonca, T., Ferreira, P.M., Marques, J.S., Marcal, A.R.S. and Rozeira, J. (2013) PH2—A Dermoscopic Image Database for Research and Benchmarking. 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, 3-7 July 2013, 5437-5440. https://doi.org/10.1109/embc.2013.6610779 |
| [26] | Tschandl, P., Rosendahl, C. and Kittler, H. (2018) The HAM10000 Dataset, a Large Collection of Multi-Source Dermatoscopic Images of Common Pigmented Skin Lesions. Scientific Data, 5, 1-9. https://doi.org/10.1038/sdata.2018.161 |
| [27] | Alom, M.Z., Hasan, M., Yakopcic, C., et al. (2024) Recurrent Residual Convolutional Neural Network Based on U-Net (R2u-Net) for Medical Image Segmentation. http://arxiv.org/abs/1802.06955 |
| [28] | Karras, T., Aittala, M., Lehtinen, J., Hellsten, J., Aila, T. and Laine, S. (2024) Analyzing and Improving the Training Dynamics of Diffusion Models. 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, 17-21 June 2024, 24174-24184. https://doi.org/10.1109/cvpr52733.2024.02282 |
