VANCOUVER, British Columbia — Artificial intelligence is dominating the tech headlines this year thanks to chatbots like ChatGPT. However, a new model developed in Canada appears to be capable of doing far more than just holding a conversation or writing someone’s term paper. Scientists from the University of British Columbia and BC Cancer have developed a new AI model that appears more accurate at predicting cancer patient survival than previous tools. Moreover, the new AI system is able to simultaneously make use of all relevant, readily available data.
The new model takes advantage of natural language processing (NLP), which is a branch of AI that focuses on and understands complex human language. This key skill allows the AI to analyze oncologist notes following a patient’s initial consultation, which is the typical first step in every cancer patient’s journey after their diagnosis. Through the identification of characteristics unique to each individual patient, the model displayed the ability to predict six-month, 36-month, and 60-month survival rates with over 80 percent accuracy.
“Predicting cancer survival is an important factor that can be used to improve cancer care,” says lead author Dr. John-Jose Nunez, a psychiatrist and clinical research fellow with the UBC Mood Disorders Centre and BC Cancer, in a university release. “It might suggest health providers make an earlier referral to support services or offer a more aggressive treatment option upfront. Our hope is that a tool like this could be used to personalize and optimize the care a patient receives right away, giving them the best outcome possible.”
How do the AI’s cancer calculations differ from a doctor’s?
Most of the time, doctors usually calculate cancer survival rates retrospectively, and then categorized according to just a few generic factors such as cancer site and tissue type. While oncologists are quite familiar with these rates, it can still be very challenging to accurately predict a certain patient’s survival due to the variety of complex factors constantly at play influencing patient outcomes.
The new AI model put together by Dr. Nunez and his team, including researchers from BC Cancer and UBC’s departments of computer science and psychiatry, can actually detect unique clues within a patient’s initial consultation document to create a more nuanced medical assessment. Even better, the model is applicable to all cancers. Many previous models are limited to certain cancer types only.
“The AI essentially reads the consultation document similar to how a human would read it,” Dr. Nunez explains. “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.”
Study authors trained and tested the AI model using data pertaining to 47,625 patients spread across all six BC Cancer sites in British Columbia. Dissimilar to chart reviews conducted by human research assistants, the new AI approach also features the added benefit of maintaining complete confidentiality when it comes to patient records.
“Because the model is trained on B.C. data, that makes it a potentially powerful tool for predicting cancer survival here in the province,” Dr. Nunez adds.
AI could become a ‘virtual assistant for physicians’
The research team believes that someday in the future this technology could help cancer clinics across Canada 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,” Dr. Nunez comments. “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.”
Dr. Nunez is a distinguished cancer researcher; he is a recipient of the 2022/23 UBC Institute of Mental Health Marshall Fellowship, and is also receives funding from the BC Cancer Foundation. In another project, he is researching how to facilitate the best-possible psychiatric and counseling care for cancer patients using advanced AI techniques.
“I see AI acting almost like a virtual assistant for physicians,” Dr. Nunez concludes. “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 improve quality of life and outcomes for patients.”
The study is published in JAMA Network Open.