|
|
关联自注意力机制改进的UNet用于肺结节图像分割
|
Abstract:
肺癌在各项肺部疾病中是非常致命的，而早期的诊断和治疗则能够有效的降低死亡率，对于早期的预防以及诊断，肺结节分割是一个非常重要的步骤，但结节与其领域之间的相似性难以处理。在这里本文提出了一个端到端的基于UNet的分割框架，称为PR-Net。为了保证标准卷积的可接受精度以及深层结构带来的信息损失，本文构建了残差结构的变体作为网络的主干。此外在解码过程中由于信息的冗余和语义差距导致难以有效的融合病变区域的空间及语义信息，为此本文采用了Transformer中的注意力机制，通过该机制的多头自注意力来有效提升空间细节和语义定位级别的特征识别。最终的模型IOU指标达到了97.71%，Dice相似系数达到了98.84%。大量的实验和结果表明，本文的模型在肺结节分割上有着非常不错的性能及稳定性。
Lung cancer is very fatal among all lung diseases, while early diagnosis and treatment can effective-ly reduce the mortality rate. For early prevention as well as diagnosis, lung nodule segmentation is a very important step, but the similarity between the nodule and its domain is difficult to handle. Here in this paper we propose an end-to-end UNet-based segmentation framework called PR-Net. To ensure acceptable accuracy of standard convolution as well as loss of information due to deep structure, a variant of residual structure is constructed as the backbone of the network. In addition, in the decoding process, the redundancy of information and semantic gaps lead to the difficulty of effectively fusing the spatial and semantic information of the lesion region, for this reason, this pa-per adopts the attention mechanism in Transformer, and effectively improves the feature recogni-tion at the spatial detail and semantic localization level through the multi-head self-attention of this mechanism. The final model IOU index reaches 97.71% and Dice similarity coefficient reaches 98.84%. A large number of experiments and results show that the model in this paper has very good performance and stability in lung nodule segmentation.
| [1] | Jalali, Y., Fateh, M., Rezvani, M., Abolghasemi, V. and Anisi, M.H. (2021) ResBCDU-Net: A Deep Learning Framework for Lung CT Image Segmentation. Sensors, 21, Article No. 268. https://doi.org/10.3390/s21010268 |
| [2] | Ding, J., Li, A., Hu, Z. and Wang, L. (2017) Accurate Pulmonary Nodule Detection in Computed Tomography Images Using Deep Convolu-tional Neural Networks. Medical Image Computing and Computer Assisted Intervention—MICCAI 2017: 20th International Conference, Quebec City, 11-13 September 2017, 559-567.
https://doi.org/10.1007/978-3-319-66179-7_64 |
| [3] | Xie, X., Pan, X., Shao, F., Zhang, W. and An, J. (2022) MCI-Net: Multi-Scale Context Integrated Network for Liver CT Image Segmentation. Computers and Electrical Engineering, 101, 108085.
https://doi.org/10.1016/j.compeleceng.2022.108085 |
| [4] | Mansoor, A., Bagci, U., Foster, B., Xu, Z., Papadakis, G.Z., Folio, L.R. and Mollura, D.J. (2015) Segmentation and Image Analysis of Abnormal Lungs at CT: Current Approaches, Chal-lenges, and Future Trends. Radiographics, 35, 1056-1076. https://doi.org/10.1148/rg.2015140232 |
| [5] | Sluimer, I., Pro-kop, M. and Van Ginneken, B. (2005) Toward Automated Segmentation of the Pathological Lung in CT. IEEE Transactions on Medical Imaging, 24, 1025-1038. https://doi.org/10.1109/TMI.2005.851757 |
| [6] | Yin, S., Deng, H., Xu, Z., Zhu, Q. and Cheng, J. (2022) SD-UNet: A Novel Segmentation Framework for CT Images of Lung Infections. Electronics, 11, 130. https://doi.org/10.3390/electronics11010130 |
| [7] | Ronneberger, O., Fischer, P. and Brox, T. (2015) U-net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015: 18th International Conference, Munich, 5-9 October 2015, 234-241. https://doi.org/10.1007/978-3-319-24574-4_28 |
| [8] | Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N. and Liang, J. (2018) UNet++: A Nested U-Net Architecture for Medical Image Segmentation. International Workshop on Multimodal Learning for Clinical Decision Support, Granada, 20 September 2018, 3-11. https://doi.org/10.1007/978-3-030-00889-5_1 |
| [9] | Milletari, F., Navab, N. and Ahmadi, S.A. (2016) V-net: Fully Con-volutional Neural Networks for Volumetric Medical Image Segmentation. 2016 4th International Conference on 3D Vision (3DV), Stanford, 25-28 October 2016, 565-571.
https://doi.org/10.1109/3DV.2016.79 |
| [10] | Hu, S., Hoffman, E.A. and Reinhardt, J.M. (2001) Automatic Lung Segmen-tation for Accurate Quantitation of Volumetric X-Ray CT Images. IEEE Transactions on Medical Imaging, 20, 490-498. https://doi.org/10.1109/42.929615 |
| [11] | Gordienko, Y., Gang, P., Hui, J., Zeng, W., Kochura, Y., Alienin, O. and Stirenko, S. (2019) Deep Learning with Lung Segmentation and Bone Shadow Exclusion Techniques for Chest X-Ray Analy-sis of Lung Cancer. International Conference on Computer Science, Engineering and Education Applications, Kiev, 26-27 January 2019, 638-647.
https://doi.org/10.1007/978-3-319-91008-6_63 |
| [12] | Rossi, F. and Rahni, A.A.A. (2015) Combination of Low Level Processing and Active Contour Techniques for Semi-Automated Volumetric Lung Lesion Segmentation from Thoracic CT Images. 2015 IEEE Student Symposium in Biomedical Engineering & Sciences (ISSBES), Shah Alam, 4 November 2015, 26-30.
https://doi.org/10.1109/ISSBES.2015.7435887 |
| [13] | Maqsood, M., Yasmin, S., Mehmood, I., Bukhari, M. and Kim, M. (2021) An Efficient DA-net Architecture for Lung Nodule Segmentation. Mathematics, 9, Article No. 1457. https://doi.org/10.3390/math9131457 |
| [14] | Luo, S., Zhang, J., Xiao, N., Qiang, Y., Li, K., Zhao, J. and Song, P. (2022) DAS-Net: A Lung Nodule Segmentation Method Based on Adaptive Dual-Branch Attention and Shadow Mapping. Applied Intelligence, 52, 15617-15631.
https://doi.org/10.1007/s10489-021-03038-2 |
| [15] | Roy, R., Chakraborti, T. and Chowdhury, A.S. (2019) A Deep Learn-ing-Shape Driven Level Set Synergism for Pulmonary Nodule Segmentation. Pattern Recognition Letters, 123, 31-38. https://doi.org/10.1016/j.patrec.2019.03.004 |
| [16] | Oktay, O., Schlemper, J., Folgoc, L.L., Lee, M., Heinrich, M., Misawa, K. and Rueckert, D. (2018) Attention u-net: Learning Where to Look for the Pancreas. |
| [17] | Paluru, N., Dayal, A., Jenssen, H.B., Sakinis, T., Cenkeramaddi, L.R., Prakash, J. and Yalavarthy, P.K. (2021) Anam-Net: Anamorphic Depth Embed-ding-Based Lightweight CNN for Segmentation of Anomalies in COVID-19 Chest CT Images. IEEE Transactions on Neural Networks and Learning Systems, 32, 932-946.
https://doi.org/10.1109/TNNLS.2021.3054746 |
| [18] | Li, Y., Chouzenoux, E., Charmettant, B., Benatsou, B., Lamarque, J.P. and Lassau, N. (2021) Lightweight u-net for Lesion Segmentation in Ultrasound Images. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), Nice, 13-16 April 2021, 611-615. https://doi.org/10.1109/ISBI48211.2021.9434086 |
| [19] | Fan, D.P., Zhou, T., Ji, G.P., Zhou, Y., Chen, G., Fu, H. and Shao, L. (2020) Inf-net: Automatic Covid-19 Lung Infection Segmentation from CT Images. IEEE Transactions on Medical Imaging, 39, 2626-2637.
https://doi.org/10.1109/TMI.2020.2996645 |
| [20] | Wang, K.N., Yang, X., Miao, J., Li, L., Yao, J., Zhou, P. and Ni, D. (2022) AWSnet: An Auto-Weighted Supervision Attention Network for Myocardial Scar and Edema Segmentation in Mul-ti-Sequence Cardiac Magnetic Resonance Images. Medical Image Analysis, 77, Article ID: 102362. https://doi.org/10.1016/j.media.2022.102362 |
