A third of cancer patients experience chronic pain, a debilitating condition that can significantly diminish quality of life, even after cancer remission. While doctors have some methods to address chronic pain, identifying those at the highest risk remains challenging. However, a new study by researchers from the University of Florida and other institutions employs artificial intelligence to predict which breast cancer patients are most likely to develop chronic pain. This predictive model could help doctors tackle underlying conditions contributing to chronic pain, leading to more effective treatments.
“We aim to understand the factors that transition a patient from having cancer to experiencing chronic pain and how we can better manage these factors,” explained Lisiane Pruinelli, Ph.D., M.S., R.N., FAMIA, the study’s senior author and a professor of family, community, and health systems science at the UF College of Nursing. “Our goal is to connect this information to patient profiles to identify early on those at risk of developing chronic pain.”
The study’s findings were published on July 26 in the Journal of Nursing Scholarship, with contributions from Pruinelli, Jung In Park, Ph.D., R.N., FAMIA, from the University of California, Irvine, and Steven Johnson, Ph.D., from the University of Minnesota.
The results indicated that with detailed data on over 1,000 breast cancer patients, the AI model could accurately predict more than 80% of the time which patients would develop chronic pain. Key factors associated with chronic pain included anxiety and depression, previous cancer diagnoses, and certain infections.
To implement a model like this in clinical settings, it would need to be integrated into the electronic healthcare records systems prevalent in clinics, requiring further research. The researchers noted that the advent of AI has the potential to help doctors customize treatments based on a patient’s unique disease characteristics.
“With the vast amount of data available and the use of artificial intelligence, we can now personalize treatments based on patient needs and their likely response to treatment,” Pruinelli said.
The study utilized extensive data from the All of Us Research Program, a nationwide initiative by the National Institutes of Health aimed at collecting anonymized healthcare records from 1 million Americans.
“This research wouldn’t be possible without individuals contributing their data,” Pruinelli added.