Program Information
Patient-Specific Qa for Cyberknife Using a Portable Flat-Panel Imager
J Gersh1,2,3*, R DuBose1 , K Shay3 , R Fulkerson4 , (1) Gibbs Cancer Center & Research Institute, Greer , SC, (2) Spectrum Medical Physics LLC, Greenville, SC, (3) Bon Secours St. Francis Cancer Center, Greenville, SC, (4) RKF Consultants LLC, Dundee, NY
Presentations
MO-DE-FS1-8 (Monday, July 31, 2017) 1:45 PM - 3:45 PM Room: Four Seasons 1
Purpose: Currently, the CyberKnife InCise MLC system delivers fields using a static-leaf configuration. Before dynamic MLC deliveries can be introduced, development of a high-resolution (spatial and temporal) detector is required for adequate patient-specific QA (PQA). This study describes a PQA acquisition of an InCise-collimated CyberKnife plan using the QA StereoChecker (QASC, Standard Imaging Inc., Middleton, WI); a high-speed, high-resolution portable flat-panel detector. While providing filmless PQA, the temporal resolution (200ms) of the QASC allows for analysis of individual beam segments and dynamic MLC plans once introduced.
Methods: An empirically-derived fluence-to-dose conversion was applied to images of individual beams with diameters of 7.5mm, 30.0mm, and 60mm. A PQA plan was generated in the treatment planning system (TPS) for an 83-segment prostate plan. Following fiducial-based automated alignment of the QASC with the CK’s target localization system, the plan was delivered from a nominal position orthogonal to the detector. Images were converted to dose and compared with dose distributions calculated by the TPS.
Results: Profile analysis of the circular fields showed that measured and planned values agreed well in the primary, shoulder, and penumbra regions of the beam. In the tail, the QASC underestimated dose, and an overcompensation algorithm was used to mitigate the over-response of higher pixel intensities resulting from photoelectron contamination originating from the amorphous silicon substrate. This overcompensation was visible when analyzing planned segments or composite images, however if analyzed with a 20% threshold, this effect was eliminated. The gamma passing percentage for the PQA, using a 2%/2mm criteria, was 98.5%. Images could be acquired every 200ms.
Conclusion: Following a dose-to-pixel value correlation, the QASC adequately and efficiently performed patient-specific QA. The high acquisition rate allowed for separate analysis of each segment, which can be applied to QA of dynamic MLC plans in the future.
Funding Support, Disclosures, and Conflict of Interest: The primary author is the owner of Spectrum Medical Physics LLC of Greenville, SC. This company provides paid consulting services for Standard Imaging Inc. of Middleton, WI. The coauthor is the owner of RKF Consults LLC of Dundee, NY. This company provides paid consulting services for Standard Imaging Inc.
Contact Email: