Room: AAPM ePoster Library
Purpose: In this study, we retrospectively investigated the magnitude of cardiac motion for a cohort of 13 patients treated with SBRT for ventricular tachycardia.
Methods: Current clinical planning follows a typical SBRT planning workflow where uniform margins are added to the electrocardiographic signal to create a PTV. For a better understanding of the cardiac motion magnitude, we employed a deformable registration to track the scar shape between imaging sets using a B-Spline deformable registration method coupled. Displacement amplitude due to respiratory\cardiac motion is visualized and tracked in 4D CTs. Setup accuracy with existing clinical workflow was determined by comparing the online rigid setup with a deformable registration between the planning CT and setup CBCT.
Results: Cardiac displacements were patient-specific with means ranging from 0.91 to 2.84 mm and maximums from 4.8 to 18 mm. Although scar points moved smoothly as denoted by the standard deviation (average 1.6 mm across all patients) displacements depended on the scar location within the heart, largest observed for a patient with an exterior left ventricle scar while smallest being recorded for an interior left ventricular lesion. When comparing the online setup with a scar tracking with deformable registration, while the average displacements were 2 mm across al patients, maximum displacements larger than 7 mm have been an observer for 7 out of 13 patients, leading to a consistent degradation of the PTV coverage for each case analyzed.
Conclusion: This study assesses the accuracy of the current planning paradigm compared to an improved workflow where the cardiac motion is tracked through deformable registration. A review of the imaging from the online setup and deformable match confirmed that the current online setup is sub-optimal for patient positioning and suggests a clinical strategy is needed to compensate for cardiac motion based on deformable registration.
Not Applicable / None Entered.