NIH's TrialGPT AI Algorithm Streamlines Clinical Trial Matching
核心洞察
The NIH has developed TrialGPT, an AI algorithm designed to accelerate matching potential volunteers to clinical trials listed on ClinicalTrials.gov (搜索).
TrialGPT identifies relevant clinical trials, explains how individuals meet enrollment criteria, and ranks trials by relevance and eligibility.
A study showed TrialGPT achieved nearly the same accuracy as clinicians in assessing patient eligibility, while reducing screening time by 40%.
Researchers at the National Institutes of Health (NIH) have unveiled TrialGPT, an artificial intelligence (AI) algorithm developed to expedite the matching of potential volunteers to clinical research trials listed on ClinicalTrials.gov (搜索). The AI is designed to help clinicians navigate the vast landscape of available clinical trials, potentially improving enrollment and accelerating medical research.
A study published in Nature Communications details how TrialGPT successfully identifies relevant clinical trials for eligible individuals and provides clear explanations of how they meet the study's enrollment criteria. The algorithm leverages large language models (LLMs) to process patient summaries containing relevant medical and demographic information. It then identifies suitable clinical trials from ClinicalTrials.gov (搜索), excludes those for which the patient is ineligible, and explains the rationale behind the eligibility assessment. The final output is an annotated list of clinical trials, ranked by relevance and eligibility, for clinicians to discuss with their patients.
"Machine learning and AI technology have held promise in matching patients with clinical trials, but their practical application across diverse populations still needed exploration," said Stephen Sherry, PhD, Acting Director of the NLM. "This study shows we can responsibly leverage AI technology so physicians can connect their patients to a relevant clinical trial that may be of interest to them with even more speed and efficiency."
Accuracy and Efficiency
To evaluate TrialGPT's accuracy, researchers compared its results to those of three human clinicians who assessed over 1,000 patient-criterion pairs. TrialGPT achieved nearly the same level of accuracy as the clinicians. A pilot user study involving two clinicians reviewing six anonymous patient summaries and matching them to clinical trials revealed that clinicians using TrialGPT spent 40% less time screening patients while maintaining the same level of accuracy.
Implications for Clinical Research
Finding the right clinical trial for interested participants is a time-consuming and resource-intensive process, which can slow down important medical research. "Our study shows that TrialGPT could help clinicians connect their patients to clinical trial opportunities more efficiently and save precious time that can be better spent on harder tasks that require human expertise," said Zhiyong Lu, PhD, NLM Senior Investigator and corresponding author of the study.
Future Directions
The research team has been selected for The Director’s Challenge Innovation Award to further assess the model’s performance and fairness in real-world clinical settings. The researchers anticipate that this work could make clinical trial recruitment more effective and help reduce barriers to participation for populations underrepresented in clinical research. The study was co-authored by collaborators from Albert Einstein College of Medicine, New York City; University of Pittsburgh; University of Illinois Urbana-Champaign; and University of Maryland, College Park.