At the University of British Columbia and BC Cancer, a research group has come up with an artificial intelligence (AI) model that has the potential to anticipate cancer patient survival more precisely and with more readily available data compared to the tools that were used earlier.
Image Credit: metamorworks/Shutterstock.com
The model makes use of natural language processing (NLP, a branch of AI that comprehends complicated human language) to examine oncologist notes after an initial consultation visit of the patient, which is the initial step in the cancer journey following diagnosis.
By determining characteristics that are unique to each patient, the model was able to forecast 6-month, 36-month, and 60-month survival with above 80% precision. The study outcomes were recently reported in the journal JAMA Network Open.
Predicting cancer survival is an important factor that can be used to improve cancer care. 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.
Dr. John-Jose Nunez, Study Lead Author and Psychiatrist, Clinical Research Fellow, Mood Disorders Centre and BC Cancer, University of British Columbia
Conventionally, cancer survival rates have been evaluated retrospectively and classified by just a few generic factors like tissue type and cancer site. In spite of familiarity with such rates, it can be hard for oncologists to precisely anticipate the survival of the individual patient as a result of the many complicated factors that impact patient outcomes.
The newly developed model by Dr. Nunez and his collaborators, including scientists from BC Cancer and UBC’s departments of computer science and psychiatry, is capable of picking up on special clues within a patient’s initial consultation document to offer a more nuanced assessment. Also, it is relevant to all cancers, whereas earlier models have been restricted to some cancer types.
The AI essentially reads the consultation document similar to how a human would read it. 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.
Dr. John-Jose Nunez, Study Lead Author and Psychiatrist, Clinical Research Fellow, Mood Disorders Centre and BC Cancer, University of British Columbia
The scientists trained and tested the model with the help of data collected from 47,625 patients throughout all six BC Cancer sites situated throughout British Columbia.
For privacy to be protected, all patient data remained stored in a safe manner at BC Cancer and was anonymously presented. Contrary to chart reviews by human research assistants, the new AI method has the added advantage of retaining the complete confidentiality of patient records.
Dr. Nunez stated, “Because the model is trained on B.C. data, that makes it a potentially powerful tool for predicting cancer survival here in the province.”
In the forthcoming days, the technology can be employed in cancer clinics throughout Canada and globally.
The great thing about neural NLP models is that they are highly scalable, portable, and don’t require structured data sets. 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. John-Jose Nunez, Study Lead Author and Psychiatrist, Clinical Research Fellow, Mood Disorders Centre and BC Cancer, University of British Columbia
Dr. Nunez is a recipient of the 2022/23 UBC Institute of Mental Health Marshall Fellowship and is also assisted by financial support from the BC Cancer Foundation. In one more stream of work, Dr Nunez is analyzing how to streamline the best possible counseling and psychiatric care for cancer patients by adopting advanced AI methods.
Nunez imagines a future where AI would be integrated into several aspects of the health system to enhance patient care.
Dr. Nunez stated, “I see AI acting almost like a virtual assistant for physicians. 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.”
Journal Reference:
Nunez, J-J., et al. (2023) Predicting the Survival of Patients With Cancer From Their Initial Oncology Consultation Document Using Natural Language Processing. JAMA Network Open. doi.org/10.1001/jamanetworkopen.2023.0813