PURPOSE: To test the potential of MRI-based treatment plan simulation for ion radiotherapy in the brain region. MATERIALS AND METHODS: A classification-based tissue segmentation method based on discriminant analysis was employed to derive so-called pseudo CT numbers from MR images of three patients with lesions in the head region undergoing ion radiotherapy. Treatment plans for ions, and for comparison purposes also for photons, were subsequently optimized and simulated using both MRI-based pseudo CT and a standard X-ray-based reference CT. RESULTS: Pseudo CTs revealed mean absolute errors in CT number in the range of 141-165 HU. While soft tissue was in good agreement with reference CT values, large deviations appeared at air cavities and bones as well as at interfaces of different tissue types. In simulations of ion treatment plans, pseudo CT optimizations showed small underdosages of target volumes with deviations in the PTV mean dose of 0.4-2.0% in comparison to reference CT optimizations. In contrast, the PTV mean dose in photon treatment plans differed by no more than 0.2%. CONCLUSIONS: The main challenge in deriving pseudo CT numbers from MRI was the correct assignment of air and compact bone. In this study, the impact of deviations on simulations of ion and photon treatment plans in the brain region was small, however for more complicated morphologies a further improvement of the classification method including MR imaging of compact bone is required.
PURPOSE: To test the potential of MRI-based treatment plan simulation for ion radiotherapy in the brain region. MATERIALS AND METHODS: A classification-based tissue segmentation method based on discriminant analysis was employed to derive so-called pseudo CT numbers from MR images of three patients with lesions in the head region undergoing ion radiotherapy. Treatment plans for ions, and for comparison purposes also for photons, were subsequently optimized and simulated using both MRI-based pseudo CT and a standard X-ray-based reference CT. RESULTS: Pseudo CTs revealed mean absolute errors in CT number in the range of 141-165 HU. While soft tissue was in good agreement with reference CT values, large deviations appeared at air cavities and bones as well as at interfaces of different tissue types. In simulations of ion treatment plans, pseudo CT optimizations showed small underdosages of target volumes with deviations in the PTV mean dose of 0.4-2.0% in comparison to reference CT optimizations. In contrast, the PTV mean dose in photon treatment plans differed by no more than 0.2%. CONCLUSIONS: The main challenge in deriving pseudo CT numbers from MRI was the correct assignment of air and compact bone. In this study, the impact of deviations on simulations of ion and photon treatment plans in the brain region was small, however for more complicated morphologies a further improvement of the classification method including MR imaging of compact bone is required.
Keywords:
Adaptive radiotherapy; Ion beam therapy; Ion radiotherapy; Magnetic resonance imaging; Photon radiotherapy; Plan simulation; Treatment planning
Authors: Shangjie Ren; Wendy Hara; Lei Wang; Mark K Buyyounouski; Quynh-Thu Le; Lei Xing; Ruijiang Li Journal: Int J Radiat Oncol Biol Phys Date: 2016-12-14 Impact factor: 7.038
Authors: Samaneh Kazemifar; Ana M Barragán Montero; Kevin Souris; Sara T Rivas; Robert Timmerman; Yang K Park; Steve Jiang; Xavier Geets; Edmond Sterpin; Amir Owrangi Journal: J Appl Clin Med Phys Date: 2020-03-26 Impact factor: 2.102
Authors: Ghazal Shafai-Erfani; Yang Lei; Yingzi Liu; Yinan Wang; Tonghe Wang; Jim Zhong; Tian Liu; Mark McDonald; Walter J Curran; Jun Zhou; Hui-Kuo Shu; Xiaofeng Yang Journal: Int J Part Ther Date: 2019-09-30