The Myeloproliferative Neoplasms Channel is supported with funding from Takeda (Gold) and Kartos Therapeutics, Inc. (Bronze).
VJHemOnc is an independent medical education platform. Supporters, including channel supporters, have no influence over the production of content. The levels of sponsorship listed are reflective of the amount of funding given to support the channel.
ASH 2025 | Using artificial intelligence to accelerate patient eligibility screening in clinical trials
In this video, Aaron Gerds, MD, Cleveland Clinic, Cleveland, OH, discusses the potential use of a medically trained large language model (LLM)-based end-to-end system to automate patient eligibility screening in clinical trials. Dr Gerds highlights the potential of this artificial intelligence (AI) system to efficiently identify eligible patients, particularly those with rare diseases, by leveraging natural language processing to screen patients across a health system, which can accelerate clinical trial enrollment and bring new treatments to patients faster. This interview took place at the 67th ASH Annual Meeting and Exposition, held in Orlando, FL.
These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.
Transcript
AI is here. AI is here to stay. It’s everywhere. And I think we need to harness its power for good. And one of the ways we can harness its power for good is by doing clinical research. One of the major barriers to enrolling patients in clinical trials is actually identifying and pre-screening those patients, particularly patients who have rarer diseases. We used polycythemia vera in our abstract, particularly for two reasons...
AI is here. AI is here to stay. It’s everywhere. And I think we need to harness its power for good. And one of the ways we can harness its power for good is by doing clinical research. One of the major barriers to enrolling patients in clinical trials is actually identifying and pre-screening those patients, particularly patients who have rarer diseases. We used polycythemia vera in our abstract, particularly for two reasons. One, it’s a rare disease, and most of the care doesn’t happen at main academic centers. It happens at regional facilities, at doctors’ offices, away from the main campuses, if you will, of academic centers. And so how do we reach those patients with a trial that’s a frontline study where they might benefit from it? Or at least we can get them on the study and see if these therapies are a benefit to them in the greater polycythemia vera population.
So what we did is we have a partner, Dionea Health, who has developed this medically trained natural language model. And we used that model, basically took the trial protocol, plugged it into the natural language model, and used it to look for patients who might be eligible for this clinical trial. And within a few minutes, we were able to identify over 30 patients who are eligible or nearly eligible for the clinical trial, just like that, right? And so just for a sense of context, you know, I’d be in my clinic normally seeing these patients. I might get a patient once a month or every other month that potentially would be eligible, and I counsel them on the trial and then I might get a patient on every three or six months. But just to plug it into a computer and it spits out 30 potential participants for a clinical trial. I mean that is efficiency at its peak and so I think this is a really powerful tool that we can use to help do trials more efficiently, more accurately, and get these newer treatments out to patients faster.
This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.
Read more...
