[an error occurred while processing this directive] | [an error occurred while processing this directive]
Improving the accuracy of pencil beam dose calculation of intensity-modulated proton therapy for lung cancer patients using deep learning
Wu Chao1,2, Pu Yuehu3, Shang Haijiao1,2
1Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China; 2University of Chinese Academy of Sciences, Beijing 100049, China; 3Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, China
AbstractObjective Proton pencil beam (PB) dose calculation can achieve rapid dose calculation, whereas it is inaccurate due to the approximation in dealing with inhomogeneities. Monte Carlo (MC) dose calculation is recognized as the most accurate method, but it is extremely time consuming. The aim of this study was to apply deep-learning methods to improve the accuracy of PB dose calculation by learning the difference between the MC and PB dose distribution. Methods A model which could convert the PB dose into the MC dose in lung cancer patients treated with intensity-modulated proton therapy (IMPT) was established based on the Hierarchically Densely Connected U-Net (HD U-Net) network. PB dose and CT images were used as model input to predict the MC dose for IMPT. The beam dose and CT images of 27 non-small cell lung cancer patients were preprocessed to the same angle and normalized, and then used as model input. The accuracy of the model was evaluated by comparing the mean square error and γ passing rate (1mm/1%) results between the predicted dose and MC dose. Results The predicted dose showed good agreement with MC dose. Using the 1mm/1% criteria, the average γ passing rate (voxels receiving more than 10% of maximum MC dose) between the predicted and MC doses reached (92.8±3.4)% for the test patients. The average dose prediction time for test patients was (6.72±2.26) s. Conclusion A deep-learning model that can accurately predict the MC dose based on the PB dose and CT images is successfully developed, which can be used as an efficient and practical tool to improve the accuracy of PB dose calculation for IMPT in lung cancer patients.
Corresponding Authors:
Pu Yuehu, Email:puyuehu@sari.ac.cn
Cite this article:
Wu Chao,Pu Yuehu,Shang Haijiao. Improving the accuracy of pencil beam dose calculation of intensity-modulated proton therapy for lung cancer patients using deep learning[J]. Chinese Journal of Radiation Oncology, 2021, 30(8): 811-816.
Wu Chao,Pu Yuehu,Shang Haijiao. Improving the accuracy of pencil beam dose calculation of intensity-modulated proton therapy for lung cancer patients using deep learning[J]. Chinese Journal of Radiation Oncology, 2021, 30(8): 811-816.
[1] 粒子治疗协作组织. 临床运行中的粒子治疗装置统计[EB/OL][2020-02-05]http://www.ptcog.ch/index.php/facilities-in-operation. [2] Paganetti H. Range uncertainties in proton therapy and the role of Monte Carlo simulations[J]. Phys Med Biol, 2012, 57(11):R99–117. DOI:10.1088/0031-9155/57/11/R99. [3] McIntosh C, Purdie TG. Voxel-based dose prediction with multi-patient atlas selection for automated radiotherapy treatment planning[J]. Phys Med Biol, 2017, 62(2):415-431. DOI:10.1088/1361-6560/62/2/415. [4] Nguyen D, Long T, Jia X,et al. A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning[J]. Sci Rep, 2019, 9(1):1076. DOI:10.1038/s41598-018-37741-x. [5] Nguyen D, Jia X, Sher D, et al. 3D radiotherapy dose prediction on head and neck cancer patients with a hierarchically densely connected U-net deep learning architecture[J]. Phys Med Biol, 2019, 64(6):065020. DOI:10.1088/1361-6560/ab039b. [6] Chen X, Men K, Li Y, et al. A feasibility study on an automated method to generate patient-specific dose distributions for radiotherapy using deep learning[J]. Med Phys, 2019, 46(1):56-64. DOI:10.1002/mp.13262. [7] Fan J, Wang J, Chen Z, et al. Automatic treatment planning based on three-dimensional dose distribution predicted from deep learning technique[J]. Med Phys, 2019, 46(1):370-381. DOI:10.1002/mp.13271. [8] Kearney V, Chan JW, Haaf S, et al. DoseNet:a volumetric dose prediction algorithm using 3D fully-convolutional neural networks[J]. Phys Med Biol, 2018, 63(23):235022. DOI:10.1088/1361-6560/aaef74. [9] Barragán-Montero AM, Nguyen D, Lu W, et al. Three-dimensional dose prediction for lung IMRT patients with deep neural networks:robust learning from heterogeneous beam configurations[J]. Med Phys, 2019, 46(8):3679-3691. DOI:10.1002/mp.13597. [10] Ronneberger O, Fischer P, Brox T. U-net:convolutional networks for biomedical image segmentation[C]// Medical image computing and computer-assisted intervention-MICCAI 2015. Munich Germany:Springer international publishing, 2015. DOI:10.1007/978-3-319-24574-4. [11] Huang G, Liu Z, Van De Maaten L, et al. Densely connected convolutional networks[C]//2017 IEEE conference on computer vision and pattern recognition (CVPR). Honolulu:IEEE computer society, 2017:2261-2269. DOI:10.1109/CVPR.2017.243. [12] Jia X, Schümann J, Paganetti H, et al. GPU-based fast Monte Carlo dose calculation for proton therapy[J]. Phys Med Biol, 2012, 57(23):7783-7797. DOI:10.1088/0031-9155/57/23/7783. [13] Hong L, Goitein M, Bucciolini M, et al. A pencil beam algorithm for proton dose calculations[J]. Phys Med Biol, 1996, 41(8):1305-1330. DOI:10.1088/0031-9155/41/8/005. [14] Schaffner B, Pedroni E, Lomax A. Dose calculation models for proton treatment planning using a dynamic beam delivery system:an attempt to include density heterogeneity effects in the analytical dose calculation[J]. Phys Med Biol, 1999, 44(1):27-41. DOI:10.1088/0031-9155/44/1/004. [15] Paganetti H. Monte Carlo simulations will change the way we treat patients with proton beams today[J]. Br J Radiol, 2014, 87(1040):20140293. DOI:10.1259/bjr.20140293. [16] Taylor PA, Kry SF, Followill DS. Pencil beam algorithms are unsuitable for proton dose calculations in lung[J]. Int J Radiat Oncol Biol Phys, 2017, 99(3):750-756. DOI:10.1016/j.ijrobp.2017.06.003. [17] Bueno M, Paganetti H, Duch M A, et al. An algorithm to assess the need for clinical Monte Carlo dose calculation for small proton therapy fields based on quantification of tissue heterogeneity[J]. Med Phys, 2013, 40(8):081704. DOI:10.1118/1.4812682. [18] Grassberger C, Daartz J, Dowdell S, et al. Quantification of proton dose calculation accuracy in the lung[J]. Int J Radiat Oncol Biol Phys, 2014, 89(2):424-430. DOI:10.1016/j.ijrobp.2014.02.023. [19] Saini J, Maes D, Egan A, et al. Dosimetric evaluation of a commercial proton spot scanning Monte-Carlo dose algorithm:comparisons against measurements and simulations[J]. Phys Med Biol, 2017, 62(19):7659-7681. DOI:10.1088/1361-6560/aa82a5. [20] Schuemann J, Giantsoudi D, Grassberger C, et al. Assessing the Clinical Impact of Approximations in Analytical Dose Calculations for Proton Therapy[J]. Int J Radiat Oncol Biol Phys, 2015, 92(5):1157-1164. DOI:10.1016/j.ijrobp.2015.04.006. [21] Liu W, Schild S E, Chang J Y, et al. Exploratory study of 4D versus 3D robust optimization in intensity-modulated proton therapy for lung cancer[J]. I Int J Radiat Oncol Biol Phys, 2016, 95(1):523-533. DOI:10.1016/j.ijrobp.2015.11.002.