Robust and efficient process for acceptance testing of radiation therapy machines

Abstract

The current (legacy) paradigm for acceptance testing (AT), commissioning, and ongoing quality assurance (QA) and quality control (QC) of linac and treatment planning systems (TPS) is extremely inefficient requiring substantial labor efforts and significant expense for the testing equipment and software. Despite these efforts and capabilities of modern radiation therapy devices the AT, commissioning and ongoing QA/QC frequently result in errors leading to large variability in implementation of these systems. This dissertation intended to develop and validate a robust, efficient process for acceptance testing of radiation therapy machines using the electronic portal imaging device (EPID). The project aimed to 1) show that the variability in EPID response can be normalized and that EPID can be used for correlating beam energy changes, 2) show that AT of a linac can be automated and can be performed using EPID as the measurement tool, and 3) compare the current paradigm of AT to the new EPID-based AT approach by performing a failure mode effects analysis (FMEA). The presented work shows that EPID would serve as an excellent tool for monitoring changes in beam energy and for performing AT of a radiation therapy machine. The FMEA results showed a significant decrease in failure pathways for the EPID-based AT approach compared to the current paradigm. The EPID-based method of QA requires minimal time and the automation that is afforded through automatic machine setup greatly reduces chances of error associated with human driven processes. The work also showed that standardizing the process could reduce the variability in implementation and also reduce the dependency on 3rd party vendor tools.