Design & Delivery of Automated Winston-Lutz Test for Isocentric & Off-Axis Delivery Stability Utilizing Truebeam Developer Mode & Electronic Portal Imaging Device

The uncertainties in treatment delivery cannot be ignored in radiation therapy. Thus, the quality assurance QA tests are very important task of the medical physicist in clinical practice. Assuring the coincidence between the mechanical isocenter of the Linear Accelerator (LINAC) and its radiation beams isocenter is one of the most important qualities need to be tested, and the Winston Lust (WL) test is the most popular technique to perform this task, especially for the treatment modalities which need high precision in beam delivery such as the stereotactic radiosurgery/stereotactic body radiotherapy (SRS/SBRT). The linear accelerator-based SRS/SBRT is a well-established method in radiation therapy. There is a recent interest in the single-isocenter technique to treat multiple lesions. However, there is a shortage in studying the accuracy of this technique, to verify the mechanical field center coincidence with the radiation field center when both are off-isocenter. In the first part of this work, an automatic WL was designed in purpose to be used in routine QA tasks. More images were acquired at broader combinations of the gantry and couch rotation angles. The 20 images automated WL needed less than 13 min to be performed, where the regular manual WL test for 8 images required an average time of 29 minutes. Also, the Automated WL only needed one-time setup and no need to go inside the treatment room between each image acquisition to change the setup, this decreased the chances of any possible errors. In the second part, an innovative Python code was developed to extract the MultiLeaves Collimator MLC positions at the cardinal angles of a conformal arc treatment plan, which was designed to treat multiple lesions located at distances 2, 4, 6, 8 cm off-isocenter, and then were exported as Digital Imaging and Communication in Medicine format (DICOM) file to Python. Out of these DICOM treatment planning files, Python would generate an eXtensible Markup Language (XML) file of the automated WL test with all the collimating leaves positions presented in the treatment planning, to acquire images at the cardinal angles for each offisocenter displacement. After feeding the generated XML files of the automated WL test to the developer mode of the True Beam LINAC, images were acquired at the cardinal angles to quantify the inaccuracy of the isocenter of the beam in targeting the center of the lesion, which was 2, 4, 6 and 8 cm off-isocenter of the LINAC machine. The resulted measurements indicate that the single isocenter multiple lesions technique complies with the recommended maximum tolerance for the LINAC-based SRS/SBRT treatment. Finally, the pitch and roll weight compensation of the six Degree of Freedom (6DoF) couch of the Varian TrueBeam LINAC was tested. The robotic couch adjusted the pitch angle to adjust the coordinate of the target vertically at every off-isocenter displacement, which made the positioning of the target in the middle of the beam possible more accurate