Room: Stars at Night Ballroom 2-3
Purpose: The goal of this research is to develop a model based on cine MRI acquired during irradiation in order to investigate the impact of internal organ motion on the cumulative dose distribution as radiation delivery parameters are varied
Methods: We develop a dose model by mapping MRI frames to 4D-CT phases. To do this, we obtain the deformation vector field (DVF) between each cine frame and a reference frame and between each 4D-CT phase and the breath-hold CT. We also map a region of interest (ROI) contoured on the breath-hold CT onto the reference cine frame. We then identify the best matching 4D-CT phase to each cine frame as the one with the most similar DVF over the ROI. The dose distribution associated with each segment of the treatment plan was calculated on each of the 4D-CT phases using a Monte Carlo dose calculation tool and was then warped to the breath-hold CT using the DVFs between 4D-CT phases and the breath-hold CT. To estimate the dose distribution deposited from each segment, we identify the sequence of best-matching 4D-CT phases to the cine frames acquired during that segment's exposure time and aggregate the corresponding warped dose distributions on the breath-hold CT.
Results: The proposed dose model is employed to quantify the potential benefit of respiratory-gated delivery using five liver cancer cases previously treated with the MRIdian system. Additionally, we investigate the sensitivity of the cumulative dose on the order in which the segments are delivered. Preliminary results for a liver case show that changes in the radiation delivery parameters could lead to minimum target dose discrepancies as large as 8%.
Conclusion: In this study, a model is proposed to aggregate and compare the dose deposited from a sequence of segments on a reference image under different radiation delivery parameters.
Funding Support, Disclosures, and Conflict of Interest: This research is funded in part by National Science Foundation (NSF) through award number no. 1662819.