AbstractObjective Hybrid attention U-net (HA-U-net) neural network was designed based on U-net for automatic delineation of craniospinal clinical target volume (CTV) and the segmentation results were compared with those of U-net automatic segmentation model. Methods The data of 110 craniospinal patients were reviewed, Among them, 80 cases were selected for the training set, 10 cases for the validation set and 20 cases for the test set. HA-U-net took U-net as the basic network architecture, double attention module was added at the input of U-net network, and attention gate module was combined in skip-connection to establish the craniospinal automatic delineation network model. The evaluation parameters included Dice similarity coefficient (DSC), Hausdorff distance (HD) and precision. Results The DSC, HD and precision of HA-U-net network were 0.901±0.041,2.77±0.29 mm and 0.903±0.038, respectively, which were better than those of U-net (all P<0.05). Conclusion The results show that HA-U-net convolutional neural network can effectively improve the accuracy of automatic segmentation of craniospinal CTV, and help doctors to improve the work efficiency and the consistent delineation of CTV.
Fund:National Natural Science Foundation of China (11775098)
Corresponding Authors:
Wang Yang,Email:Janetcyj@163.com
Cite this article:
Li Hongwei,Ni Chunxia,Chen Shu et al. Automatic delineation of craniospinal clinical target volume based on hybrid attention U-net[J]. Chinese Journal of Radiation Oncology, 2022, 31(3): 266-271.
Li Hongwei,Ni Chunxia,Chen Shu et al. Automatic delineation of craniospinal clinical target volume based on hybrid attention U-net[J]. Chinese Journal of Radiation Oncology, 2022, 31(3): 266-271.
[1] Gajjar A, Chintagumpala M, Ashley D, et al. Risk-adapted craniospinal radiotherapy followed by high-dose chemotherapy andstem-cell rescue in children with newly diagnosedmedulloblastoma (Si Jude Medulloblastoma-96):long-term results from a prospective, multicentre trial[J]. Lancet Oncol, 2006, 7(10):813-820. DOI:10.1016/S1470-2045(06) 70867-1. [2] Packer RJ, Gajjar A, Vezina G, et al. Phase Ⅲ study of craniospinal radiation therapy followed by adjuvant chemotherapy for newly diagnosed average-risk medulloblastoma[J]. J Clin Oncol, 2006, 24(25):4202-4208. DOI:10. 1200/JCO.2006.06.4980. [3] Noble DJ, Scoffings D, Ajithkumar T, et al. Fast imaging employing steady-state acquisition (FIESTA) MRI to investigate cerebrospinal fluid (CSF) within dural reflections of posterior fossa cranial nerves[J]. Br J Radiol, 2016, 89(1067):20160392. DOI:10.1259/bjr.20160392. [4] Ajithkumar T, Horan G. Padovani L, et al. SIOPE-Brain tumor group consensus guideline on craniospinal target volume delineation for high-precision radiotherapy[J]. Radiother Oncol, 2018, 128(2):192-197. DOI:10.1016/j.ra-donc.2018.04.016. [5] 邢鹏飞,杨咏强,钱建军,等. 儿童全脑全脊髓放疗SIOPE指南全脑靶区勾画临床应用研究[J]. 中华放射肿瘤学杂志,2020, 29(4):262-266. DOI:10.3760/cma.j.cn.113030-20190805-00003. Xing PF, Yang YQ, Qian JJ, et al. Clinical application of SIOPE guidelines in target definition for craniospinal irradiation in children[J]. Chin J Radiat Oncol, 2020, 29(4):262-266. DOI:10.3760/cma. j.cn.113030-20190805-00003. [6] Men K, Geng HZ, Cheng CY, et al. Technical note:more accurate and efficient segmentation of organs-at-risk in radiotherapy with convolutional neural networks cascades[J]. Med Phys, 2019, 46(1):286-292. DOI:10.1002/mp. 13296. [7] Fu Y, Mazur TR, Wu X, et al. A novel MRI segmentation method using CNN based correction network for MRI guided adaptive radiotherapy[J]. Med Phys, 2018, 45(11):5129-5137. DOI:10.1002/mp.13221. [8] 葛迦,宁丽华,严森详,等. 两种软件Smart Segmentation与MIM Altas自动勾画鼻咽癌危及器官的准确性研究[J]. 中华放射医学与防护杂志,2019, 39(9):668-672. DOI:10.3760/cma.j.issn.0254-5098.2019. 09.006. Ge J, Ning LH, Yan SX, et al. Automatic segmentation of organs at risk for nasopharyngeal carcinoma with Smart Segmentation and MIM Atlas[J]. Chin J Radiol Med Protect, 2019, 39(9):668-672. DOI:10.3760/cma. j.issn.0254-5098.2019. 09.006. [9] 陈辛元,门阔,唐玉,等. 基于深度学习自动分割模型乳腺癌放疗临床应用与评价[J]. 中华放射肿瘤学杂志,.2020, 29(3):197-202. DOI:10.3760/c ma.j.issn. 1004-4221. 2020.03.009. Chen XY, Men K, Tang Y, et al. Clinical application and evaluation of automatic segmentation model based on deep learning for breast cancer radiotherapy[J]. Chin J Radiat Oncol, 2020, 29(3):197-202. DOI:10.3760/c ma.j. issn. 1004-4221.2020.03.009. [10] Ibragimov B, Xing L. Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks[J]. Med Phys, 2017, 44(2):547-557. DOI:10.1002/mp.12045. [11] Ronneberger O,Fischer P, Brox T. U-net:convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-assisted Intervention.Cham:Springer, 2015:234-241. DOI:10. 1007/978-3-319-24574-4_28. [12] Oktay O, Schlemper J, Folgoc LL, et al. Attention U-net:learning where to look for the pancreas[EB/OL][2021-01-10].https://arxiv.org/pdf/1804.03999. [13] Fu J, Liu J,Tian HJ,et al. Dual attention network for scene segmentation[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach:IEEE,2019:3146-3154. DOI:10.1109/CVPR.2019.0326. [14] Woo S, Park J, Lee JY, et al. CBAM:convolutional block attention module[C]//European Conference on Computer Vision. Berlin:Springer, 2018:3-19. DOI:10.1007/978-3-030-01234-2_1. [15] 郝晓宇,熊俊峰,薛旭东,等. 融合双注意力机制3D U-net的肺肿瘤分割[J]. 中国图象图形学报,2020, 25(10):2119-2127. DOI:10.11834/jig.200 282. Hao XY, Xiong JF, Xue XD, et al.3D U-net with dual attention mechanism for lung tumor segmentation[J]. Chin J Image Graph, 2020, 25(10):2119-2127. DOI:10.11834/jig.200282. [16] Crum WR, Camara O, Hill DLG. Generalized overlap measures for evaluation and validation in medical image analysis[J]. IEEE Trans Med Imaging, 2006, 25(11):1451-1461. DOI:10. 1109/TMI. 2006. 880587. [17] 张倩雯,陈明,秦玉芳,等. 基于3D ResU-net网络的肺结节分割[J]. 中国医学物理学杂志,2019, 36(11):1356-1361. DOI:10.3969/j.issn.1005-202X. 2019.11. 021. Zhang QW, Chen M, Qin YF, et al. Lung nodule segmentation based on 3D ResU-net network[J]. Chin J Med Phys, 2019, 36(11):1356-1361. DOI:10.3969j.issn. 1005-202X.2019. 11.021. [18] Huttenlocher DP, Klanderman GA, Rucklidge WJ. Comparing images using the Hausdorff distance[J]. IEEE Trans Pattern Anal Mach Intell, 1993, 15(9):850-863. DOI:10. 1109/ 34. 232073. [19] 秦楠楠,薛旭东,吴爱林,等. 基于U-net卷积神经网络的宫颈癌临床靶区和危及器官自动勾画的研究[J]. 中国医学物理学杂志,2020, 37(4):524-528. DOI:10.3969/j.issn. 1005-202X.2020.04.023. Qin NN, Xue XD, Wu AL, et al. Automatic segmentation of clinical target volumes and organs-at-risk in radiotherapy for cervical cancer using U-net convo-lutionalneural network[J]. Chin J Med Phys, 2020, 37(4):524-528. DOI:10.3969/j.issn.1005-202X.2020.04.023.
2022, Vol. 31 
