AI reduces time needed to generate radiation therapy plans
Researchers from the University of Toronto have developed an artificial intelligence tool that needs less time to create treatment plans for cancer patients.
While it can often take days to produce radiation therapy plans tailored for each patient, engineers at UT have cut the time down to just hours to generate such individualized treatment plans.
Aaron Babier, an engineering researcher, and his team at UT’s Department of Mechanical and Industrial Engineering, developed the automated software that leverages AI to mine historical radiation therapy data, which is then applied to an optimization engine to come up with radiation therapy plans.
“Head and neck cancer can be particularly challenging because of the locations of the targeted tumors in relation to healthy tissue,” according to Babier, who says the algorithm is customized for each patient. “Obviously, in a cancer treatment plan, we normally want to deliver a minimum radiation dose to healthy tissue. That will vary from patient to patient.”
In a study of 217 patients with throat cancer, the AI-based software achieved comparable results to conventionally planned treatments in as little as 20 minutes, based on results recently published in the journal Medical Physics.
“There have been other AI optimization engines that have been developed—the idea behind ours is that it more closely mimics the current clinical best practice,” notes Babier. “Right now, treatment planners have this big time sink. If we can intelligently burn this time sink, they’ll be able to focus on other aspects of treatment. The idea of having automation and streamlining jobs will help make healthcare costs more efficient.”
Ultimately, Babier sees the software serving as a clinical decision support tool providing automated radiation therapy treatment planning. However, he adds that any treatment plan produced by the tool would still need to be reviewed and further customized by a radiation physicist.
“A human being will always be involved,” Babier concludes. “At the end of the day, a patient is always going to want a physician to sign off on the plan.”