Geometric evaluation and quantifying dosimetric impact of diverse deformable image registration algorithms on abdomen images with biomechanically modeled deformations

Purpose:Deformable image registration (DIR) has been increasingly used inradiation therapy (RT). The accuracy of DIR algorithms and how it impacts onthe RT plan dosimetrically were examined in our study for abdominal sites usingbiomechanically modeled deformations. Methods:Five pancreatic cancer patients were enrolled in this study. Follow-ing the guidelines of AAPM TG-132, a patient-specific quality assurance (QA)workflow was developed to evaluate DIR for the abdomen using the TG-132recommended virtual simulation software ImSimQA (Shrewsbury,UK).First,theplanning CT was deformed to simulate respiratory motion using the embeddedbiomechanical model in ImSimQA. Additionally, 5 mm translational motion wasadded to the stomach, duodenum, and small bowel. The original planning CTand the deformed CT were then imported into Eclipse and MIM to perform DIR.The output displacement vector fields (DVFs) were compared with the groundtruth from ImSimQA. Furthermore, the original treatment plan was recalculatedon the ground-truth deformed CT and the deformed CT (with Eclipse and MIMDVF). The dose errors were calculated on a voxel-to-voxel basis. Results:Data analysis comparing DVF from Eclipse versus MIM show theaverage mean DVF magnitude errors of 2.8 +/- 1.0 versus 1.1 +/- 0.7 mm forstomach and duodenum, 5.2 +/- 4.0 versus 2.5 +/- 1.0 mm for small bowel, and4.8 +/- 4.1 versus 2.7 +/- 1.1 mm for the gross tumor volume (GTV), respec-tively, across all patients. The mean dose error on stomach+duodenum andsmall bowel were 2.3 +/- 0.6% for Eclipse, and 1.0 +/- 0.3% for MIM. As the DIRmagnitude error increases, the dose error range increase, for both Eclipse andMIM. Conclusion:In our study, an initial assessment was conducted to evaluate theaccuracy of DIR and its dosimetric impact on radiotherapy. A patient-specificDIR QA workflow was developed for pancreatic cancer patients. This workflowexhibits promising potential for future implementation as a clinical workflow.