Artificial intelligence used to predict cancer patient survival 'more accurately': B.C. research

A team of researchers out of the University of British Columbia and BC Cancer has developed an artificial intelligence tool to better predict a cancer patient's survival rate by reading their doctor's notes.

In a news release Thursday, UBC said this new model helps predict survival rates "more accurately" and with "more readily available data" than previous tools.

"Predicting cancer survival is an important factor that can be used to improve cancer care," said lead author Dr. John-Jose Nunez in the release.

Nunez, a psychiatrist and clinical research fellow at the UBC Mood Disorders Centre and BC Cancer, said the AI could suggest an earlier referral for support services or a more aggressive treatment plan.

"Our hope is that a tool like this could be used to personalize and optimize the care a patient receives right away," Nunez said, adding that it could help give patients the best outcome possible.

Researchers said the model uses natural language processing — a branch of AI that understands complex human language — to analyze an oncologist's notes after a patient's consultation.

The results show that the model was able to predict six-month, 36-month and 60-month survival with greater than 80 per cent accuracy by identifying characteristics unique to each patient, UBC said.

"The AI essentially reads the consultation document similar to how a human would read it," said Nunez. "These documents have many details like the age of the patient, the type of cancer, underlying health conditions, past substance use, and family histories. The AI brings all of this together to paint a more complete picture of patient outcomes."

Nunez added that this new tool is applicable to all cancers, where previous models have been limited to certain types of cancer.

Researchers tested the model using data from 47,625 patients from six BC Cancer locations across B.C.

"Because the model is trained on B.C. data, that makes it a potentially powerful tool for predicting cancer survival here in the province," said Nunez.

He said he hopes the technology could someday be used in cancer clinics across the country and around the world.

"The great thing about neural NLP models is that they are highly scalable, portable and don't require structured data sets," he said. "We can quickly train these models using local data to improve performance in a new region. I would suspect that these models provide a good foundation anywhere in the world where patients are able to see an oncologist."

Nunez, a recipient of the 2022-23 UBC Institute of Mental Health Marshall Fellowship, is also working on how to facilitate the best possible psychiatric and counselling care for cancer patients using AI techniques.

"I see AI acting almost like a virtual assistant for physicians," he said. "As medicine gets more and more advanced, having AI to help sort through and make sense of all the data will help inform physician decisions. Ultimately, this will help imrpove quality of life and outcomes for patients."