Program Information
Evaluation of Scatter Correction with Modified Scatter Kernel Model for CBCT-Based Radiotherapy
Y Nomura1*, Q Xu2,3 , H Peng2,3 , H Shirato2,4 , S Shimizu2,5 , L Xing2,3 , (1) Department of Radiation Oncology, Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaido, (2) Global Station for Quantum Medical Science and Engineering, Global Institution for Collaborative Research and Education (GI-CoRE), Hokkaido University, Sapporo, Hokkaido, (3) Department of Radiation Oncology, Stanford University, Palo Alto, CA (4) Department of Radiation Medicine, Faculty of Medicine and Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaido, (5) Department of Radiation Oncology, Faculty of Medicine and Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaido
Presentations
MO-L-GePD-J(A)-5 (Monday, July 31, 2017) 1:15 PM - 1:45 PM Room: Joint Imaging-Therapy ePoster Lounge - A
Purpose: Electron density (ED) has played a crucial role in radiation therapy treatment planning system (TPS). Although cone-beam computed tomography (CBCT) can calculate patient ED images just before radiotherapy, image quality is degraded due to scatter. This study aims to demonstrate scatter correction using modified scatter kernel model (MSKM) and evaluate its effectiveness for CBCT-based ED images and dose distributions in radiotherapy.
Methods: First, scatter kernel model was modified by considering both glare and constant value shift effects. Iterative deconvolution algorithm was applied to extract primary projections with these modified kernels. Following this step, Hounsfield unit images were reconstructed by Feld-kamp Davis Kress (FDK) and then converted to ED images by look-up tables. Finally, intensity-modulated photon therapy (IMRT) and intensity-modulated proton therapy (IMPT) plans were generated with the ED images to evaluate dose distribution errors compared to reference ED image. The proposed method was performed on Monte Carlo simulated data with virtual CIRS and Gammex phantoms.
Results: The primary intensity was accurately estimated from simulated projections with the MSKM. Reconstructed image contrast, cupping artefacts and pixel value compression were improved. Root mean squared ED errors compared to reference values decreased from 3.47 % to 0.81 %. Dose distribution errors were observed around the edge of phantom in deconvolved IMRT plan. The mean dose and the dose received over 95 % of planning target volume in simulated IMPT plan were overestimated by 19.0 % and 43.6 % respectively whereas those in deconvolved plan were -1.84 % and -1.53 %.
Conclusion: This study revealed the MSKM-corrected ED images had superior results to radiotherapy treatment planning. IMPT plans were more sensitive to ED image accuracy than IMRT and significantly overestimated doses. This correction technique can be utilized in image-guided radiotherapy and offline dose verifications.
Contact Email: