Grant: Collaborative Health Research Projects (NSERC partnered)

The delivery of radiation for the treatment of cancer is a complicated process that requires both clinical and technical expertise to ensure radiation treatments are safe and effective. Sub-optimal treatment plans have the potential to result in significant detriment to the patient and several studies have shown radiation therapy (RT) plans, which deviate from established quality assurance (QA) guidelines, result in worse patient outcomes. Therefore, the current RT process requires substantial multi-disciplinary QA resources to reduce the likelihood of errors and to ensure a high standard of patient care. Technical innovations in RT have improved the quality of treatment plans usually at the cost of increased complexity. Therefore, more information and plan data requires review on the part of the multi-disciplinary RT team. The goal of the current research proposal is to integrate automation into the existing QA process in order to improve the quality of the RT and to develop a system for reducing errors. In order to achieve these objectives, we are developing sophisticated mathematical algorithms, which represent and store thousands of previous RT plans deemed to be safe and of high quality. For a new radiation treatment plan created, the algorithms examine features of the plan to establish relationships and patterns in the data in order to identify deviations and errors in comparison to the plan knowledge base. These algorithms, once integrated into an automated QA system will simplify the plan review process and provide the RT team essential knowledge to make more informed decisions about patient treatment. This research will also make the RT process more efficient and robust, thereby reducing interruptions in patient treatments and reducing wait times for patients requiring RT. The research will promote widespread dissemination and knowledge translation to other institutions, thus benefiting all patients receiving RT as part of their cancer management.