The kVCT System Commissioning of a Novel Medical Linear Accelerator Designed for Biology-Guided Radiotherapy.

PURPOSE/OBJECTIVE(S) A novel therapeutic machine that combines PET, fan beam kVCT and linear accelerator to support real-time PET-guided treatment, termed biology-guided radiotherapy (BgRT). This study reports the positioning accuracy, image quality, and dose commissioning of the kVCT imaging system of the first clinical BgRT machine at our institute. MATERIALS/METHODS The helical fan-beam kVCT subsystem consists of an X-ray tube with a voltage of 120kV and a 16-row GOS Ceramic Scintillator detector. The gantry rotation speed is 60 rpm. The tube currents are 45, 67, 133mA for 0.5 pitch; 100, 150, and 300mA for 0.5 pitch; 267 and 400mA for 1.333 pitch. The image slice thickness is 1.25mm with a field of view of 500mm and a resolution of 512 × 512. A ball-cube phantom was used to test the positioning accuracy of the kVCT subsystem with respect to the setup laser and MV beam. Catphan504 phantom was imaged with 0.5 pitch and 150mAs to test the kVCT image quality of the BgRT system, including the uniformity, noise, HU constancy, spatial resolution, geometric distortion, slice thickness, and low contrast visibility and detectability using the PIPSproTM software. The RTI MAS-2 and Piranha detectors were used to verify the accuracy of the mA and voltage. The CTDI dose was measured using a CTDI phantom and a Standard Imaging A101 pencil chamber. RESULTS The localization error of the ball cube phantom was -0.8, 0.9, and 0.3mm in IEC x, y, and z directions, respectively. For the Catphan504 tests, the vertical and horizontal uniformities were 0.926 and 0.857, respectively. The HU constancy tests of the air, water, polystyrene, acrylic, Delrin, Teflon were all within 40HU to the expected values. The modulation transfer function test results were 0.467, 0.528, and 0.595 lp/mm for F50, F40, and F30, respectively, which showed a comparable spatial resolution to regular CT simulators. The geometric distortion was 0.12mm and slice thickness inaccuracy was 0.13mm. The low contrast detectability was 0.734. The measured mA and voltage were within 5% of the nominal setting of 120 kV and 150 mA. The CTDIw was less than 1.4 cGy for body and 3 cGy for head with a protocol of 120 kVp, 150 mAs, and 1.25 mm slice thickness. CONCLUSION This evaluation represents the kVCT characteristics of the novel BgRT system with architecture designed to accommodate CT, PET and a LINAC. The image quality and HU constancy are comparable to our CT simulators and is a useful tool for online adaptive radiotherapy. The commissioning data will serve as a standard for units installed at future clinical sites.