Roupen Odabashian, Oncologist at Abbotsford Regional Hospital and Cancer Centre, Founder of MeDucation AI, Podcast Host at OncoDaily, shared on LinkedIn:
“Fewer than one in twenty adults with cancer enrolls in a clinical trial. A randomized study this year tested whether AI could change that, at the least glamorous step: prescreening charts for eligibility.
Trained research staff screened 355 patients with lung or colorectal cancer alone, versus staff augmented with a language model. The AI arm improved accuracy on the hardest criteria to abstract, biomarkers and tumor details, while preserving speed. The gains were modest. They were also real, and they showed up exactly where humans struggle most: reading dense unstructured notes for a dozen inclusion criteria at once.
This is the part of oncology AI that does not trend. Not a model diagnosing cancer, just a model reading the chart well enough to notice a patient might qualify for a trial nobody flagged. The upside is not efficiency. It is that a patient who would have been missed gets an option they did not have.
If the bottleneck to trial enrollment is attention rather than eligibility, that is a problem AI is unusually well suited to attack.”
Title: Human-AI teaming to improve accuracy and efficiency of eligibility criteria prescreening for oncology trials: a randomized evaluation trial using retrospective electronic health records
Authors: Ravi B. Parikh, Likhitha Kolla, Elizabeth A. Beothy, William J. Ferrell, Brenda Laventure, Matthew Guido, Anthony Girard, Yang Li, Khaled Essam Mahmoud Dosoky, Karim Tarabishy, Parth S. Patel, Ayana Andalcio, Kristin Maloney, Jose Ulises Mena, Wael Salloum, Jinbo Chen, Ezekiel J. Emanuel
Read the Full Article.
Other articles featuring Roupen Odabashian on OncoDaily.