Artificial Intelligence Improves Radiation Therapy at RVH
BARRIE - Radiation therapy, or radiotherapy, is an essential part of treatment plans for many cancer patients. The goal is to destroy or damage cancer cells by using high doses of radiation without hurting nearby healthy cells; however, before treatment can begin, the borders between tumours and adjacent healthy organs must be clearly identified by radiation oncologists in a process known as contouring.
With the incidence of cancer rising with aging populations, there is a growing need to optimize and automate different radiotherapy tasks to improve efficiency and help reduce physician workload. Researchers at the Simcoe Muskoka Regional Cancer Program led a recent study showing that deep learning technology, a form of artificial intelligence that uses complex algorithms to identify patterns, can help improve the efficiency of the prostate cancer contouring process.
The study, published in the journal Practical Radiation Oncology, and in collaboration with McMaster University and the University of Toronto, used deep-learning software to accurately define the relevant borders and organ volumes in a fraction of the time normally required by manual contouring.
RVH physicist and lead author Dr. Nevin McVicar says “historically, radiation therapy teams have been reluctant to adopt fully automated contouring processes, but our study provides a promising step forward for using artificial intelligence to improve patient care”.