Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN
PurposeRecent deep-learning based synthetic computed tomography (sCT) generation using magnetic resonance (MR) images have shown promising results. However, generating sCT for the abdominal region poses challenges due to the patient motion, including respiration and peristalsis. To address these cha...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2024.1478148/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841561074744164352 |
|---|---|
| author | Chanwoong Lee Chanwoong Lee Young Hun Yoon Young Hun Yoon Young Hun Yoon Jiwon Sung Jun Won Kim Yeona Cho Jihun Kim Jaehee Chun Jin Sung Kim Jin Sung Kim Jin Sung Kim |
| author_facet | Chanwoong Lee Chanwoong Lee Young Hun Yoon Young Hun Yoon Young Hun Yoon Jiwon Sung Jun Won Kim Yeona Cho Jihun Kim Jaehee Chun Jin Sung Kim Jin Sung Kim Jin Sung Kim |
| author_sort | Chanwoong Lee |
| collection | DOAJ |
| description | PurposeRecent deep-learning based synthetic computed tomography (sCT) generation using magnetic resonance (MR) images have shown promising results. However, generating sCT for the abdominal region poses challenges due to the patient motion, including respiration and peristalsis. To address these challenges, this study investigated an unsupervised learning approach using a transformer-based cycle-GAN with structure-preserving loss for abdominal cancer patients.MethodA total of 120 T2 MR images scanned by 1.5 T Unity MR-Linac and their corresponding CT images for abdominal cancer patient were collected. Patient data were aligned using rigid registration. The study employed a cycle-GAN architecture, incorporating the modified Swin-UNETR as a generator. Modality-independent neighborhood descriptor (MIND) loss was used for geometric consistency. Image quality was compared between sCT and planning CT, using metrics including mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structure similarity index measure (SSIM) and Kullback-Leibler (KL) divergence. Dosimetric evaluation was evaluated between sCT and planning CT, using gamma analysis and relative dose volume histogram differences for each organ-at-risks, utilizing treatment plan. A comparison study was conducted between original, Swin-UNETR-only, MIND-only, and proposed cycle-GAN.ResultsThe MAE, PSNR, SSIM and KL divergence of original cycle-GAN and proposed method were 86.1 HU, 26.48 dB, 0.828, 0.448 and 79.52 HU, 27.05 dB, 0.845, 0.230, respectively. The MAE and PSNR were statistically significant. The global gamma passing rates of the proposed method at 1%/1 mm, 2%/2 mm, and 3%/3 mm were 86.1 ± 5.9%, 97.1 ± 2.7%, and 98.9 ± 1.0%, respectively.ConclusionThe proposed method significantly improves image metric of sCT for the abdomen patients than original cycle-GAN. Local gamma analysis was slightly higher for proposed method. This study showed the improvement of sCT using transformer and structure preserving loss even with the complex anatomy of the abdomen. |
| format | Article |
| id | doaj-art-e57e9fd1654a43dbb5c6dbef692957df |
| institution | Kabale University |
| issn | 2234-943X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj-art-e57e9fd1654a43dbb5c6dbef692957df2025-01-03T06:46:49ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-01-011410.3389/fonc.2024.14781481478148Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GANChanwoong Lee0Chanwoong Lee1Young Hun Yoon2Young Hun Yoon3Young Hun Yoon4Jiwon Sung5Jun Won Kim6Yeona Cho7Jihun Kim8Jaehee Chun9Jin Sung Kim10Jin Sung Kim11Jin Sung Kim12Department of Radiation Oncology, Yonsei Cancer Center, Heavy Ion Therapy Research Institute, Yonsei University College of Medicine, Seoul, Republic of KoreaMedical Physics and Biomedical Engineering Lab (MPBEL), Yonsei University College of Medicine, Seoul, Republic of KoreaDepartment of Radiation Oncology, Yonsei Cancer Center, Heavy Ion Therapy Research Institute, Yonsei University College of Medicine, Seoul, Republic of KoreaMedical Physics and Biomedical Engineering Lab (MPBEL), Yonsei University College of Medicine, Seoul, Republic of KoreaDepartment of Radiation Oncology, Washington University in St. Louis, St Louis, MO, United StatesDepartment of Radiation Oncology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of KoreaDepartment of Radiation Oncology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of KoreaDepartment of Radiation Oncology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of KoreaDepartment of Radiation Oncology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of KoreaOncosoft Inc., Seoul, Republic of KoreaDepartment of Radiation Oncology, Yonsei Cancer Center, Heavy Ion Therapy Research Institute, Yonsei University College of Medicine, Seoul, Republic of KoreaMedical Physics and Biomedical Engineering Lab (MPBEL), Yonsei University College of Medicine, Seoul, Republic of KoreaOncosoft Inc., Seoul, Republic of KoreaPurposeRecent deep-learning based synthetic computed tomography (sCT) generation using magnetic resonance (MR) images have shown promising results. However, generating sCT for the abdominal region poses challenges due to the patient motion, including respiration and peristalsis. To address these challenges, this study investigated an unsupervised learning approach using a transformer-based cycle-GAN with structure-preserving loss for abdominal cancer patients.MethodA total of 120 T2 MR images scanned by 1.5 T Unity MR-Linac and their corresponding CT images for abdominal cancer patient were collected. Patient data were aligned using rigid registration. The study employed a cycle-GAN architecture, incorporating the modified Swin-UNETR as a generator. Modality-independent neighborhood descriptor (MIND) loss was used for geometric consistency. Image quality was compared between sCT and planning CT, using metrics including mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structure similarity index measure (SSIM) and Kullback-Leibler (KL) divergence. Dosimetric evaluation was evaluated between sCT and planning CT, using gamma analysis and relative dose volume histogram differences for each organ-at-risks, utilizing treatment plan. A comparison study was conducted between original, Swin-UNETR-only, MIND-only, and proposed cycle-GAN.ResultsThe MAE, PSNR, SSIM and KL divergence of original cycle-GAN and proposed method were 86.1 HU, 26.48 dB, 0.828, 0.448 and 79.52 HU, 27.05 dB, 0.845, 0.230, respectively. The MAE and PSNR were statistically significant. The global gamma passing rates of the proposed method at 1%/1 mm, 2%/2 mm, and 3%/3 mm were 86.1 ± 5.9%, 97.1 ± 2.7%, and 98.9 ± 1.0%, respectively.ConclusionThe proposed method significantly improves image metric of sCT for the abdomen patients than original cycle-GAN. Local gamma analysis was slightly higher for proposed method. This study showed the improvement of sCT using transformer and structure preserving loss even with the complex anatomy of the abdomen.https://www.frontiersin.org/articles/10.3389/fonc.2024.1478148/fullMR-linacabdominal synthetic CTstructure consistency losstransformerunsupervised learning |
| spellingShingle | Chanwoong Lee Chanwoong Lee Young Hun Yoon Young Hun Yoon Young Hun Yoon Jiwon Sung Jun Won Kim Yeona Cho Jihun Kim Jaehee Chun Jin Sung Kim Jin Sung Kim Jin Sung Kim Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN Frontiers in Oncology MR-linac abdominal synthetic CT structure consistency loss transformer unsupervised learning |
| title | Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN |
| title_full | Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN |
| title_fullStr | Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN |
| title_full_unstemmed | Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN |
| title_short | Abdominal synthetic CT generation for MR-only radiotherapy using structure-conserving loss and transformer-based cycle-GAN |
| title_sort | abdominal synthetic ct generation for mr only radiotherapy using structure conserving loss and transformer based cycle gan |
| topic | MR-linac abdominal synthetic CT structure consistency loss transformer unsupervised learning |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2024.1478148/full |
| work_keys_str_mv | AT chanwoonglee abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT chanwoonglee abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT younghunyoon abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT younghunyoon abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT younghunyoon abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jiwonsung abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT junwonkim abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT yeonacho abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jihunkim abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jaeheechun abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jinsungkim abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jinsungkim abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan AT jinsungkim abdominalsyntheticctgenerationformronlyradiotherapyusingstructureconservinglossandtransformerbasedcyclegan |