Artificial Intelligence changing the game for radiation treatment at the Princess Margaret


The Princess Margaret’s Dr. Thomas Purdie and his team, including Dr. Chris McIntosh, have used artificial intelligence (AI) to rapidly and accurately develop radiation therapy plans in minutes opposed to hours to days. This technology uses machine learning that takes information from a database of peer-reviewed and approved radiation therapy plans from  Princess Margaret Cancer Centre.

Although there are no two patients with identical conditions, there are similarities. Based on these similarities, AI technologies can use a patient’s images and find parallel features within the database, which helps build a personalized radiation therapy plan. The system is constantly and rapidly adapting to optimize plans to suit new patients, without requiring additional user input during this process.

"The technology allows radiation medicine teams to take on more complex cases and provide precision medicine to more patients," says Dr. Purdie. 

Last month, the AI technology moved from just research to the clinic, which impacts patient care. The clinical team is led by Dr. Alejandro Berlin, a Princess Margaret radiation oncologist, who is appling the AI technology to treat a prostate cancer patient.

"The AI plan was deemed superior for that specific patient, and got the 'green light' after meeting all our protocol and quality assurance metrics," says Dr. Berlin. "To the best of our knowledge, this is the first time anywhere that a patient has been actually treated with a radiation plan based only on AI technology.

"We are working on quantifying the impact and value of this AI method, as we we expect it will become the new paradigm of care for patients in the future."

Read the full story on the UHN Website.