NIH-developed AI algorithm matches potential volunteers to clinical trials
Researchers at the NIH have developed an AI algorithm called TrialGPT to streamline matching volunteers to relevant clinical trials on ClinicalTrials.gov. Published in Nature Communications, TrialGPT uses large language models to process patient summaries and identify suitable trials, enhancing enrollment and accelerating medical research. TrialGPT matches patients to trials by evaluating eligibility and providing summaries on why they qualify. Testing against human clinicians showed TrialGPT achieved similar accuracy but reduced screening time by 40%. This tool helps clinicians efficiently connect patients with trials, improving research progress. The NIH team, including collaborators from several universities, received The Director’s Challenge Innovation Award to further evaluate TrialGPT’s real-world performance and fairness, aiming to enhance clinical trial recruitment, especially for underrepresented populations.
For more details see the original article.
Last Updated on December 4, 2024