[Comparison of setup accuracy between ExacTrac X-ray 6 dimensions and cone-beam computed tomography for intracranial and pelvic image-guided radiotherapy].

The aim of this study was to compare the setup difference measured with ExacTrac X-ray 6D (ETX6D) and cone-beam computed tomography (CBCT) for non-invasive fractionated radiotherapy. Setup data were collected on a Novalis Tx treatment unit for both a head phantom and patients with intracranial tumors and a pelvic phantom and patients with prostate cancer. Initially, setup was done for a phantom using ETX6D. Secondly, a treatment couch was shifted or rotated by each already known value. Thirdly, ETX6D and CBCT scans were obtained. Finally, setup difference was determined: the registrations of ETX6D images with the corresponding digitally reconstructed radiographs using ETX6D fusion, and registrations of CBCT images with the planning CT using online 6D fusion. The setup difference between ETX6D and CBCT was compared. The impact of shifts and rotations on the difference was evaluated. Patients' setup data was similarly analyzed. In phantom experiments, the root mean square (RMS) of difference of the shift and rotation was less than 0.45 mm for translations, and 0.17 degrees for rotations. In intracranial patients' data, the RMS of that was 0.55 mm and 0.44 degree, respectively. In prostate cancer patients' data, the RMS of that was 0.77 mm and 0.79 degree, respectively. In this study, we observed modest setup differences between ETX6D and CBCT. These differences were generally less than 1.00 mm for translations, and 1.00 degrees for rotations, respectively.