Researchers at the National Institutes of Health are using gigantic language models to develop an artificial intelligence framework to streamline the clinical trial matching process and more quickly connect potential volunteers with appropriate trials listed on ClinicalTrials.gov.
By comparing its accuracy with three clinicians, researchers found that the TrialGPT tool achieved nearly the same level of accuracy as shown in this month’s NIH announcement.
WHY IT’S IMPORTANT
Because finding the right clinical trial for a patient takes both time and resources, researchers at the National Library of Medicine and the National Cancer Institute developed the TrialGPT platform to streamline it.
The novel clinical trial matching algorithm analyzes patient summaries for relevant medical and demographic information, then identifies clinical trials for which the patient is eligible and excludes trials for which he or she is not.
TrialGPT creates an annotated list of clinical trials – organized by suitability and eligibility – that physicians can utilize to discuss clinical trial options with their patients. The AI tool also explains how an individual meets the study’s inclusion criteria, which is crucial to its effectiveness.
To assess how well TrialGPT predicted whether a patient met specific clinical trial requirements, researchers compared the tool’s results with those of three physicians who assessed more than 1,000 patient-criterion pairs, the NIH reports.
“Machine learning and artificial intelligence technology have shown promise in matching patients to clinical trials, but their practical application in diverse populations still requires research,” Stephen Sherry, acting director of NLM, said in a statement.
The researchers also conducted a pilot user study and found that clinicians using TrialGPT spent 40% less time screening patients while maintaining the same level of accuracy.
For the study, published in Nature Communications and co-authored by collaborators from the Albert Einstein College of Medicine, the University of Pittsburgh, the University of Illinois Urbana-Champaign and the University of Maryland at College Park, the research team received an innovation award and will be further evaluated by the NIH. model performance and fairness in real clinical settings.
A BIGGER TREND
The utilize of AI to improve patient recruitment, retention, and clinical trial outcomes began before OpenAI launched its ChatGPT generative AI model. During the COVID-19 pandemic, cancer organizations have been looking for ways to find patients across the country who would be eligible for trials, even if they weren’t physically there, based on health care records.
According to Jeff Elton, CEO of ConcertAI, a provider of data and AI SaaS platforms for optimizing clinical trials, the increased utilize of artificial intelligence has helped drive decentralized clinical trials.
“With integrated digital trials, clinical trials are an integral part of the care process itself, rather than being imposed on it,” Elton said.
“Research need not impose a greater burden on providers and patients than standard care.”
According to Seth Howard, vice president of research and development at Epic, reducing friction throughout the clinical trial lifecycle is critical to ensuring patients have access to trial therapies.
An electronic health record provider has implemented a data-driven clinical trial pairing two years ago. Using its de-identified Cosmos dataset, Epic enables providers who enroll in the service to match clinical trial opportunities from sponsors and their organization’s number of eligible patients.
Many health systems have also conducted testing using analytics applications that can reveal clinical trial opportunities to patients using their organization’s EHR data. In October, Microsoft announced novel AI tools that will enable healthcare systems to create custom AI tools for many administrative needs, including clinical trial matching.
However, AI bias still raises concerns about clinical outcomes.
It can emerge at any stage of algorithm development and widen health care disparities, researchers at the Yale School of Medicine found in a novel study published earlier this month.
ON RECORDING
“Our study shows that TrialGPT can help clinicians more effectively provide clinical trial opportunities to their patients and save valuable time that can be better spent on more difficult tasks requiring human expertise,” said Zhiyong Lu, NLM senior researcher and corresponding author of the study. statement.
“This study shows that we can responsibly use artificial intelligence technology so that doctors can connect their patients even more quickly and effectively with relevant clinical trials that may be of interest to them,” Sherry added.