AI Transforming Clinical Trials for Heart Failure

Published Date: 17/06/2024

Mass General Brigham researchers have found that generative AI can rapidly and accurately screen patients for clinical trial eligibility, but safeguards are needed to ensure safety and equity.

A recent study by Mass General Brigham researchers has demonstrated the potential of generative artificial intelligence (Gen AI) to revolutionize clinical trials by rapidly and accurately screening patients for eligibility. The study, published in NEJM AI, found that a Gen AI process called RECTIFIER (RAG-Enabled Clinical Trial Infrastructure for Inclusion Exclusion Review) could identify patients who meet criteria for enrollment in a heart failure trial based on their medical records with an accuracy rate of 97.9% to 100%. This is significantly higher than the accuracy rate of disease-trained research coordinators, who typically conduct screening.


The researchers tested the ability of the AI process to identify patients eligible for the COPILOT-HF trial, which recruits patients with symptomatic heart failure and identifies potential participants based on electronic health record (EHR) data. They designed 13 prompts to assess clinical trial eligibility and tested them using medical charts of a small group of patients before applying them to a dataset of 1,894 patients with an average of 120 notes per patient.


The results of the study suggest that Gen AI could fundamentally improve clinical trial screening, making it faster and cheaper to evaluate new treatments and ultimately help bring successful ones to patients. However, the researchers noted that AI can have risks that should be monitored when it is integrated into routine workflows, such as introducing bias and missing nuances in medical notes.


The study's authors recommended that any study using AI to screen patients should have checks in place to ensure safety and equity. Most trials have a clinician who double-checks the participants who are deemed eligible for a trial, and the researchers suggested that this final check should continue with AI screening.


The researchers' goal is to prove that this technology works in other disease areas and use cases while expanding beyond the walls of Mass General Brigham.


Mass General Brigham is a non-profit academic healthcare system that consists of Massachusetts General Hospital and Brigham and Women's Hospital. The Accelerator for Clinical Transformation is a program within Mass General Brigham that aims to accelerate the translation of innovative ideas into clinical practice.

  

Mass General Brigham is a leader in medical research and innovation, with a strong commitment to improving patient care and advancing medical knowledge.

FAQs:

Q: What is the purpose of clinical trial screening?

A: Clinical trial screening is the process of identifying patients who meet specific criteria for enrollment in a clinical trial.


Q: How does Gen AI improve clinical trial screening?

A: Gen AI can rapidly and accurately screen patients for clinical trial eligibility, making it faster and cheaper to evaluate new treatments.


Q: What are the risks of using AI in clinical trials?

A: The risks of using AI in clinical trials include introducing bias and missing nuances in medical notes.


Q: How can the risks of using AI in clinical trials be mitigated?

A: The risks of using AI in clinical trials can be mitigated by having checks in place, such as a clinician who double-checks the participants who are deemed eligible for a trial.


Q: What is the goal of the researchers who conducted the study?

A: The goal of the researchers is to prove that this technology works in other disease areas and use cases while expanding beyond the walls of Mass General Brigham.

More Topics: