NIH Develops AI Algorithm 'TrialGPT' to Streamline Clinical Trial Matching

NIH Develops AI Algorithm to Revolutionize Clinical Trial Matching

Researchers from the National Institutes of Health (NIH) have developed an innovative artificial intelligence (AI) algorithm called TrialGPT, designed to streamline the process of matching potential volunteers to relevant clinical research trials listed on ClinicalTrials.gov. The study, published in Nature Communications, demonstrates the algorithm's ability to successfully identify suitable clinical trials for eligible participants and provide clear explanations of how individuals meet study enrollment criteria 1

1

2

3

How TrialGPT Works

TrialGPT, developed by a team from NIH's National Library of Medicine (NLM) and National Cancer Institute, leverages the power of large language models (LLMs) to create an efficient clinical trial matching process. The algorithm operates as follows:

1. Processes a patient summary containing relevant medical and demographic information
2. Identifies relevant clinical trials from ClinicalTrials.gov for which the patient is eligible
3. Excludes trials for which the patient is ineligible
4. Explains how the person meets the study enrollment criteria
5. Produces an annotated list of clinical trials, ranked by relevance and eligibility

This output enables clinicians to discuss clinical trial opportunities with their patients more effectively 1

1

2

3

Impressive Performance and Time Savings

To evaluate TrialGPT's performance, researchers conducted two key assessments:

1. Comparison with human clinicians: TrialGPT's results were compared to those of three human clinicians who assessed over 1,000 patient-criterion pairs. The AI algorithm achieved nearly the same level of accuracy as the clinicians 1

   1

   2

   2

   3

   3

   .

2. Pilot user study: Two human clinicians reviewed six anonymous patient summaries and matched them to six clinical trials. One clinician manually reviewed the summaries, while the other used TrialGPT. The study found that clinicians using TrialGPT spent 40% less time screening patients while maintaining the same level of accuracy 1

   1

   2

   2

   3

   3

   .

Related Stories

AI-Guided Patient Stratification Revolutionizes Alzheimer's Drug Trials

Drugmakers deploy AI to accelerate clinical trials and slash regulatory submission timelines

AI Tool Developed to Enhance Medical Study Reporting

AI Shows Promise in Clinical Decision-Making, But Challenges Remain

Potential Impact on Medical Research

The development of TrialGPT holds significant promise for accelerating medical research and improving clinical trial enrollment. Dr. Stephen Sherry, NLM Acting Director, emphasized the responsible leveraging of AI technology to connect patients with relevant clinical trials more efficiently 1

1

2

3

Dr. Zhiyong Lu, NLM Senior Investigator and corresponding author of the study, highlighted the potential time-saving benefits: "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" 1

1

2

3

Future Developments and Implications

Given the promising results, the research team has been selected for The Director's Challenge Innovation Award to further assess TrialGPT's performance and fairness in real-world clinical settings. Researchers anticipate that this work could enhance clinical trial recruitment effectiveness and help reduce participation barriers for underrepresented populations in clinical research 1

1

2

3

The study was co-authored by collaborators from Albert Einstein College of Medicine, University of Pittsburgh, University of Illinois Urbana-Champaign, and University of Maryland, College Park 1

1

2

3