AI software cuts long radiation therapy planning process to just 20 minutes
A team at the University of Toronto has successfully developed artificial intelligence (AI) that helps automate the radiation therapy planning process, potentially saving radiologists from several days of work on just one patient.
Radiation therapy planning is essential to the cancer treatment process, lead author Aaron Babier, an engineering researcher at U of T, and colleagues in the department of mechanical and industrial engineering, wrote in a recently published Medical Physics study. But the painstaking work can take time—up to several days to craft an individualized plan for each patient—and cancer care is inherently urgent.
To circumvent the lengthy process, Babier et al. developed an AI software that can mine historical radiation therapy data. Data are then applied to an optimization engine, which churns out treatment plans.
The researchers tested the tech on 217 patients with throat cancer, all of whom also received treatment using conventional radiotherapy treatment (RT) planning methods. The outcomes of therapies suggested by Babier and colleagues’ AI system were comparable to those of traditional therapy plans—but Babier’s technology could achieve those same results within 20 minutes.
“Right now treatment planners have this big time sink,” Babier said in a U of T release. “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. I think it’ll really help to ensure high-quality care.”
AI could play a big role in relieving clinicians of tedious work like RT planning and help them spend that time refining processes robots can’t help with, like improving patient care at the bedside. While artificial intelligence might help streamline workflow at hospitals, Babier said, it can’t replace trained doctors or specialists, who will still be needed to fine-tune any plans generated by the AI software.
“There have been other AI optimization engines that have been developed,” he said. “The idea behind ours is that it more closely mimics the current clinical best practice.”
After graduating from Indiana University-Bloomington with a bachelor’s in journalism, Anicka joined TriMed’s Chicago team in 2017 covering cardiology. Close to her heart is long-form journalism, Pilot G-2 pens, dark chocolate and her dog Harper Lee.