Estimation of Optimal Margin for Intrafraction Movements During Frameless Brain Radiosurgery

Purpose/Objective(s): To verify the geometric accuracy of a respiratory gating system using gated KV-CBCT and gated MV-CBCT. Materials/Methods: A commercial breathing phantom embedded with fiducial markers was positioned on the treatment couch of a linear accelerator using the gated marker match method. KV images were taken using the on-board-imager of the linear accelerator triggered by the real-time position management system. Forty gated KV images were taken at various gantry angles during the whole delivery process. The images were then used to reconstruct a KV-CBCT image using filtered back projection algorithm by an in-house developed software. The same process was repeated using the MV portal imager and a gated MV-CBCT image was reconstructed. The 3D distances between the centroids of the fiducial markers of the gated CBCT images and the planned phase of the 4D planning CT were measured using a commercial treatment planning system. The whole process was repeated 10 times with various breathing amplitudes and periods of the breathing phantom. Results: All the 3D distances measured between the centroids of the fiducial markers in the gated CBCT images and the planning CT image were less than 1.5 mm. The average values for the gated KV-CBCT images and the gated MV-CBCT images were 0.4 mm and 0.6 mm respectively. Conclusions: Gated CBCT images obtained using KV or MV imaging modality of a linear accelerator could be useful for the verification of the 3D geometric accuracy of a respiratory gating system. Author Disclosure: M.L. Cheung: None. L.K. Lee: None. M.K. Kam: None. B.K. Yu: None. A.T. Chan: None.