Dose and position quality assurance using the RADPOS system for 4D radiotherapy with cyberknife

Abstract

The CyberKnife system consists of a compact LINAC mounted on a mobile robotic arm and the Synchrony Respiratory Motion Tracking System. This complex radiotherapy system needs independent performance verification to assure safe treatments. In this work, we use the RADPOS 4D dosimetry system to verify CyberKnife’s motion tracking and delivered dose. RADPOS motion measurements are compared with internal metal fiducial and external LED marker log files. Dose measurements are compared with film and treatment planning system (TPS) calculations. RADPOS and EBT3 Gaf- Chromic films were calibrated in Solid Water (5 cm depth, 80 cm source-detector distance, 60 mm cone, Exradin ion chamber, Fluke electrometer). A CT-based treatment plan was created for a Solid Water breast phantom containing fiducials and RADPOS. Dose calculations were performed using Multi- Plan TPS, Monte Carlo (MC) and ray tracing (RT) algorithms. Before treatment, film was inserted inside the breast phantom adjacent to RADPOS. The breast phantom and LED markers were positioned on the chest platform of a Quasar Motion Phantom. Position logging began for RADPOS and Synchrony, Quasar motion started, and irradiation commenced. A coordinate alignment algorithm was implemented, allowing position tracking modalities to be compared in a common coordinate system. The average standard deviation of the differences between LED and RADPOS position measurements was 0.33, 0.39, and 0.56 mm along the left/right, superior/inferior, and anterior/posterior directions, respectively. Dose percent difference values during static phantom irradiations were 0.3% (RADPOS/RT), 1.3% (RADPOS/MC), 0.5% (Film/RT), 1.5% (Film/MC), and -0.2% (RADPOS/Film), while values during dynamic phantom irradiations were 2.0% (RADPOS/RT), 2.9% (RADPOS/MC), -1.0% (Film/RT), 0.0% (Film/MC), and 3.0% (RADPOS/Film). Average gamma results were greater than 96% for MC and RT dose calculation algorithms, and for dynamic and static treatments. Our work demonstrates that RADPOS is a useful tool for independent QA of CyberKnife treatments with Synchrony respiratory compensation.