*Result*: Unsupervised deep learning-based 4DCT deformable image registration and carbon-ion 4D dose calculation.
Title:
Unsupervised deep learning-based 4DCT deformable image registration and carbon-ion 4D dose calculation.
Authors:
An X; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China.; School of Future Technology, Xi'an Jiaotong University, Xi'an, China., Wang J; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China., Zhang J; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China., Liu Y; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; School of Future Technology, Xi'an Jiaotong University, Xi'an, China., Abuduxiku M; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China., Chai N; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China., Luo Y; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China., Wu W; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China., Hu Y; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; School of Future Technology, Xi'an Jiaotong University, Xi'an, China., Hei D; School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China., Li Q; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China.; Putian Lanhai Nuclear Medicine Research Center, Putian, China., He P; Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Heavy Ion Radiation Biology and Medicine of Chinese Academy of Sciences, Lanzhou, China.; Key Laboratory of Basic Research on Heavy Ion Radiation Application in Medicine, Lanzhou, Gansu Province, China.; University of Chinese Academy of Sciences, Beijing, China.; Putian Lanhai Nuclear Medicine Research Center, Putian, China.
Source:
Medical physics [Med Phys] 2026 Mar; Vol. 53 (3), pp. e70387.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: John Wiley and Sons, Inc Country of Publication: United States NLM ID: 0425746 Publication Model: Print Cited Medium: Internet ISSN: 2473-4209 (Electronic) Linking ISSN: 00942405 NLM ISO Abbreviation: Med Phys Subsets: MEDLINE
Imprint Name(s):
Publication: 2017- : Hoboken, NJ : John Wiley and Sons, Inc.
Original Publication: Lancaster, Pa., Published for the American Assn. of Physicists in Medicine by the American Institute of Physics.
Original Publication: Lancaster, Pa., Published for the American Assn. of Physicists in Medicine by the American Institute of Physics.
MeSH Terms:
Radiotherapy Planning, Computer-Assisted*/methods , Image Processing, Computer-Assisted*/methods , Four-Dimensional Computed Tomography* , Deep Learning* , Heavy Ion Radiotherapy* , Radiation Dosage* , Unsupervised Machine Learning*, Lung Neoplasms/diagnostic imaging ; Lung Neoplasms/radiotherapy ; Humans ; Radiotherapy Dosage ; Radiometry
References:
Bert C, Grözinger SO, Rietzel E. Quantification of interplay effects of scanned particle beams and moving targets. Phys Med Biol. 2008;53(9):2253‐2265. doi:10.1088/0031‐9155/53/9/003.
Xiao H, Xue X, Zhu M, et al. Deep learning‐based lung image registration: a review. Comput Biol Med. 2023;165:107434. doi:10.1016/j.compbiomed.2023.107434.
Ichikawa T, Kumazaki T. 4D‐CT: a new development in three‐dimensional hepatic computed tomography. J Nippon Med Sch. 2000;67(1):24‐27. doi:10.1272/jnms.67.24.
Long L, Xue X, Xiao H. CCMNet: cross‐scale correlation‐aware mapping network for 3D lung CT image registration. Comput Biol Med. 2024;182:109103. doi:10.1016/j.compbiomed.2024.109103.
Li H, Dong L, Bert C, et al. AAPM Task Group Report 290: respiratory motion management for particle therapy. Med Phys. 2022;49(4):e50‐e81. doi:10.1002/mp.15470.
Pakela JM, Knopf A, Dong L, Rucinski A, Zou W. Management of motion and anatomical variations in charged particle therapy: past, present, and into the future. Front Oncol. 2022;12:806153. doi:10.3389/fonc.2022.806153.
Keall PJ, El Naqa I, Fast MF, et al. Real‐time dose‐guided radiation therapy. Int J Radiat Oncol Biol Phys. 2025;122(4):787‐801. doi:10.1016/j.ijrobp.2025.04.019.
Gu X, Dong B, Wang J, et al. A contour‐guided deformable image registration algorithm for adaptive radiotherapy. Phys Med Biol. 2013;58(6):1889‐1901. doi:10.1088/0031‐9155/58/6/1889.
Andersen ES, Noe KØ, Sørensen TS, et al. Simple DVH parameter addition as compared to deformable registration for bladder dose accumulation in cervix cancer brachytherapy. Radiother Oncol. 2013;107(1):52‐57. doi:10.1016/j.radonc.2013.01.013.
De Silva T, Uneri A, Ketcha MD, et al. 3D‐2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch. Phys Med Biol. 2016;61(8):3009‐3025. doi:10.1088/0031‐9155/61/8/3009.
Fu YB, Chui CK, Teo CL, Kobayashi E. Motion tracking and strain map computation for quasi‐static magnetic resonance elastography. Med Image Comput Comput Assist Interv. 2011;14(pt 1):428‐435. doi:10.1007/978‐3‐642‐23623‐5_54.
Balakrishnan G, Zhao A, Sabuncu MR, Guttag J, Dalca AV. VoxelMorph: a learning framework for deformable medical image registration. IEEE Trans Med Imaging. 2019;38:1788‐1800. doi:10.1109/TMI.2019.2897538.
Zhang Y, Liu C, Liu M, et al. Attention is all you need: utilizing attention in AI‐enabled drug discovery. Brief Bioinform. 2023;25(1):bbad467. doi:10.1093/bib/bbad467.
Chen Z, Zheng Y, Gee JC. TransMatch: a Transformer‐Based Multilevel Dual‐Stream Feature Matching Network for Unsupervised Deformable Image Registration. IEEE Trans Med Imaging. 2024;43(1):15‐27. doi:10.1109/TMI.2023.3288136.
Lei Y, Fu Y, Wang T, et al. 4D‐CT deformable image registration using multiscale unsupervised deep learning. Phys Med Biol. 2020;65(8):085003. doi:10.1088/1361‐6560/ab79c4.
Antony R, Lonski P, Ungureanu E, et al. Independent review of 4DCT scans used for SABR treatment planning. J Appl Clin Med Phys. 2020;21(3):62‐67. doi:10.1002/acm2.12825.
Shackleford J, Kandasamy N, Sharp G. Chapter 6 – Plastimatch—An open‐source software for radiotherapy imaging. In: Shackleford J, Kandasamy N, Sharp G, eds. High Performance Deformable Image Registration Algorithms for Manycore Processors. Morgan Kaufmann; 2013:107‐114. doi:10.1016/B978‐0‐12‐407741‐6.00006‐2.
Chang Q, Lu C, Li M. Cascading affine and B‐spline registration method for large deformation registration of lung X‐rays. J Digit Imaging. 2023;36(3):1262‐1278. doi:10.1007/s10278‐022‐00763‐z.
He P, Li Q. Impact of different synchrotron flattop operation modes on 4d dosimetric uncertainties for scanned carbon‐ion beam delivery. Front Oncol. 2022;12:806742. doi:10.3389/fonc.2022.806742.
He P, Mori S. Perturbation analysis of 4D dose distribution for scanned carbon‐ion beam radiotherapy. Phys Med. 2020;74:74‐82. doi:10.1016/j.ejmp.2020.05.003.
Minohara S, Kanai T, Endo M, Noda K, Kanazawa M. Respiratory gated irradiation system for heavy‐ion radiotherapy. Int J Radiat Oncol Biol Phys. 2000;47(4):1097‐1103. doi:10.1016/s0360‐3016(00)00524‐1.
Zhang J, An X, Chai N, Li Q, He P. A novel uniform‐spaced fast rescanning method for carbon‐ion radiotherapy. Med Phys. 2025;52(7):e17994. doi:10.1002/mp.17994.
Brock KK, Mutic S, McNutt TR, Li H, Kessler ML. Use of image registration and fusion algorithms and techniques in radiotherapy: report of the AAPM Radiation Therapy Committee Task Group No. 132. Med Phys. 2017;44(7):e43‐e76. doi:10.1002/mp.12256.
Paganelli C, Meschini G, Molinelli S, Riboldi M, Baroni G. Patient‐specific validation of deformable image registration in radiation therapy: overview and caveats. Med Phys. 2018;45(10):e908‐e922. doi:10.1002/mp.13162.
Isensee F, Rokuss M, Krämer L, et al. nnInteractive: redefining 3D promptable segmentation. arXiv [preprint]. 2025. doi:10.48550/arXiv.2503.08373.
Xiao H, Xue X, Zhu M, et al. Deep learning‐based lung image registration: a review. Comput Biol Med. 2023;165:107434. doi:10.1016/j.compbiomed.2023.107434.
Ichikawa T, Kumazaki T. 4D‐CT: a new development in three‐dimensional hepatic computed tomography. J Nippon Med Sch. 2000;67(1):24‐27. doi:10.1272/jnms.67.24.
Long L, Xue X, Xiao H. CCMNet: cross‐scale correlation‐aware mapping network for 3D lung CT image registration. Comput Biol Med. 2024;182:109103. doi:10.1016/j.compbiomed.2024.109103.
Li H, Dong L, Bert C, et al. AAPM Task Group Report 290: respiratory motion management for particle therapy. Med Phys. 2022;49(4):e50‐e81. doi:10.1002/mp.15470.
Pakela JM, Knopf A, Dong L, Rucinski A, Zou W. Management of motion and anatomical variations in charged particle therapy: past, present, and into the future. Front Oncol. 2022;12:806153. doi:10.3389/fonc.2022.806153.
Keall PJ, El Naqa I, Fast MF, et al. Real‐time dose‐guided radiation therapy. Int J Radiat Oncol Biol Phys. 2025;122(4):787‐801. doi:10.1016/j.ijrobp.2025.04.019.
Gu X, Dong B, Wang J, et al. A contour‐guided deformable image registration algorithm for adaptive radiotherapy. Phys Med Biol. 2013;58(6):1889‐1901. doi:10.1088/0031‐9155/58/6/1889.
Andersen ES, Noe KØ, Sørensen TS, et al. Simple DVH parameter addition as compared to deformable registration for bladder dose accumulation in cervix cancer brachytherapy. Radiother Oncol. 2013;107(1):52‐57. doi:10.1016/j.radonc.2013.01.013.
De Silva T, Uneri A, Ketcha MD, et al. 3D‐2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch. Phys Med Biol. 2016;61(8):3009‐3025. doi:10.1088/0031‐9155/61/8/3009.
Fu YB, Chui CK, Teo CL, Kobayashi E. Motion tracking and strain map computation for quasi‐static magnetic resonance elastography. Med Image Comput Comput Assist Interv. 2011;14(pt 1):428‐435. doi:10.1007/978‐3‐642‐23623‐5_54.
Balakrishnan G, Zhao A, Sabuncu MR, Guttag J, Dalca AV. VoxelMorph: a learning framework for deformable medical image registration. IEEE Trans Med Imaging. 2019;38:1788‐1800. doi:10.1109/TMI.2019.2897538.
Zhang Y, Liu C, Liu M, et al. Attention is all you need: utilizing attention in AI‐enabled drug discovery. Brief Bioinform. 2023;25(1):bbad467. doi:10.1093/bib/bbad467.
Chen Z, Zheng Y, Gee JC. TransMatch: a Transformer‐Based Multilevel Dual‐Stream Feature Matching Network for Unsupervised Deformable Image Registration. IEEE Trans Med Imaging. 2024;43(1):15‐27. doi:10.1109/TMI.2023.3288136.
Lei Y, Fu Y, Wang T, et al. 4D‐CT deformable image registration using multiscale unsupervised deep learning. Phys Med Biol. 2020;65(8):085003. doi:10.1088/1361‐6560/ab79c4.
Antony R, Lonski P, Ungureanu E, et al. Independent review of 4DCT scans used for SABR treatment planning. J Appl Clin Med Phys. 2020;21(3):62‐67. doi:10.1002/acm2.12825.
Shackleford J, Kandasamy N, Sharp G. Chapter 6 – Plastimatch—An open‐source software for radiotherapy imaging. In: Shackleford J, Kandasamy N, Sharp G, eds. High Performance Deformable Image Registration Algorithms for Manycore Processors. Morgan Kaufmann; 2013:107‐114. doi:10.1016/B978‐0‐12‐407741‐6.00006‐2.
Chang Q, Lu C, Li M. Cascading affine and B‐spline registration method for large deformation registration of lung X‐rays. J Digit Imaging. 2023;36(3):1262‐1278. doi:10.1007/s10278‐022‐00763‐z.
He P, Li Q. Impact of different synchrotron flattop operation modes on 4d dosimetric uncertainties for scanned carbon‐ion beam delivery. Front Oncol. 2022;12:806742. doi:10.3389/fonc.2022.806742.
He P, Mori S. Perturbation analysis of 4D dose distribution for scanned carbon‐ion beam radiotherapy. Phys Med. 2020;74:74‐82. doi:10.1016/j.ejmp.2020.05.003.
Minohara S, Kanai T, Endo M, Noda K, Kanazawa M. Respiratory gated irradiation system for heavy‐ion radiotherapy. Int J Radiat Oncol Biol Phys. 2000;47(4):1097‐1103. doi:10.1016/s0360‐3016(00)00524‐1.
Zhang J, An X, Chai N, Li Q, He P. A novel uniform‐spaced fast rescanning method for carbon‐ion radiotherapy. Med Phys. 2025;52(7):e17994. doi:10.1002/mp.17994.
Brock KK, Mutic S, McNutt TR, Li H, Kessler ML. Use of image registration and fusion algorithms and techniques in radiotherapy: report of the AAPM Radiation Therapy Committee Task Group No. 132. Med Phys. 2017;44(7):e43‐e76. doi:10.1002/mp.12256.
Paganelli C, Meschini G, Molinelli S, Riboldi M, Baroni G. Patient‐specific validation of deformable image registration in radiation therapy: overview and caveats. Med Phys. 2018;45(10):e908‐e922. doi:10.1002/mp.13162.
Isensee F, Rokuss M, Krämer L, et al. nnInteractive: redefining 3D promptable segmentation. arXiv [preprint]. 2025. doi:10.48550/arXiv.2503.08373.
Grant Information:
12375354 National Natural Science Foundation of China; Y2023113 Youth Innovation Promotion Association, Chinese Academy of Sciences; 22JR5RA125 Natural Science Foundation of Gansu Province; 23YFFA0010 Natural Science Foundation of Gansu Province
Contributed Indexing:
Keywords: 4DCT; carbon‐ion 4D dose calculation; deformable image registration; unsupervised deep learning
Entry Date(s):
Date Created: 20260315 Date Completed: 20260316 Latest Revision: 20260315
Update Code:
20260316
DOI:
10.1002/mp.70387
PMID:
41833529
Database:
MEDLINE
*Further Information*
*Background: Accurate four-dimensional dose calculation (4DDC) is essential for carbon-ion lung radiotherapy and relies on deformable image registration (DIR). However, conventional DIR methods are computationally intensive, hindering the implementation of online adaptive workflows.
Purpose: This study investigates the feasibility and efficacy of unsupervised deep learning-based DIR models, TransMatch and VoxelMorph, in accelerating clinical lung four-dimensional computed tomography (4DCT) registration and facilitating accurate carbon-ion 4D dose calculation.
Methods: A total of 150 clinical lung 4DCT datasets were utilized (120 for training, 20 for validation, and 10 for testing), with a conventional B-spline method serving as the baseline. Registration accuracy was evaluated using the Mean Absolute Error (MAE), Dice Similarity Coefficient (DSC), 95th percentile Hausdorff Distance (HD95), and Jacobian determinant (|J|). Carbon-ion 4D dose distributions were accumulated using the generated deformation vector fields (DVFs). Dosimetric impacts on the gross tumor volume (GTV) and organs at risk (OARs) were quantified using Dose-Volume Histogram (DVH) metrics under a gating with 6× rescanning scenario.
Results: TransMatch and VoxelMorph achieved superior registration accuracy with lower MAE, mean DSC > 0.97, and HD95 < 2.5 mm. In DVF analysis, TransMatch and VoxelMorph showed negligible folding rates (<0.02%) and significantly higher Jacobian standard deviations than the B-spline method, indicating superior capability in capturing fine local deformations. Dosimetrically, differences in GTV and OARs metrics between the deep learning and B-spline methods were less than 2% of the prescription dose, falling within clinically acceptable tolerances. Crucially, TransMatch and VoxelMorph models achieved sub-second registration times (<1 s), whereas the conventional B-spline method required more than 10 min.
Conclusions: TransMatch and VoxelMorph achieve geometric and dosimetric accuracy comparable to the conventional B-spline method for carbon-ion lung radiotherapy while offering substantially higher computational speed, highlighting their potential for real-time adaptive carbon-ion therapy.
(© 2026 American Association of Physicists in Medicine.)*