Tackling Patient Recruitment Challenges from Two Perspectives with AI: TrialGPT and BEKplatform

Artificial intelligence (AI) is revolutionizing the landscape of clinical trials, particularly in the realms of patient recruitment and matching. Innovations like the National Institutes of Health’s (NIH) TrialGPT and BEKhealth’s AI-powered BEKplatform are at the forefront of this transformation, each addressing distinct facets of the recruitment process. When utilized together, these technologies have the potential to enhance patient access to clinical trials and provide researchers with more qualified participant pools.

TrialGPT: Enhancing Clinical Trial Matching for Healthcare Providers

Developed by researchers at NIH’s National Library of Medicine (NLM) and National Cancer Institute (NCI), TrialGPT is an AI algorithm designed to streamline the process of matching potential volunteers to relevant clinical research trials listed on ClinicalTrials.gov. By processing patient summaries that include medical and demographic information, TrialGPT can identify suitable clinical trials for which a patient is eligible and provides explanations on how the individual meets the study enrollment criteria. This tool promises to assist clinicians in efficiently navigating the extensive array of clinical trials available, thereby facilitating informed discussions with patients about potential participation.

In evaluations, TrialGPT demonstrated an accuracy rate of 87.3% in predicting patient eligibility, closely aligning with human expert performance. Additionally, a pilot user study revealed that clinicians using TrialGPT reduced screening time by 40% without compromising accuracy. While not being widely used yet, these findings suggest that TrialGPT can save clinicians time and accelerate clinical enrollment and research, and more research is underway to further assess the model’s performance.

BEKplatform: Empowering Clinical Researchers in Patient Identification

BEKhealth’s BEKplatform is an AI-driven patient-matching solution that enables clinical researchers to identify eligible participants more efficiently. By extracting and analyzing both structured and unstructured data from electronic medical records (EMRs), BEKplatform creates a comprehensive, longitudinal patient graph. This allows researchers to build robust queries and patient cohorts, optimizing feasibility assessments and expediting site selection.

The platform’s ability to process unstructured data, such as physician notes and lab reports, enhances its precision in matching patients to complex trial criteria. BEKhealth reports that their technology can identify ten times more qualified patients and double enrollment rates, significantly reducing the time and resources traditionally required for patient recruitment.

Complementary Roles of TrialGPT and BEKplatform

While both TrialGPT and BEKplatform leverage AI to improve the clinical trial process, they serve complementary roles within the ecosystem:

• TrialGPT: Assists healthcare providers by analyzing patient data to identify suitable clinical trials from a vast database. This facilitates informed discussions between clinicians and patients about trial participation, potentially increasing patient access to clinical trials as a care option.
• BEKplatform: Empowers clinical researchers by automating the identification of eligible patients within their own healthcare systems. By efficiently sifting through EMR data, it enables researchers to quickly assemble qualified participant pools, thereby accelerating study initiation and enhancing enrollment outcomes.

The integration of AI solutions on both the clinical and research fronts creates a synergistic effect that benefits the entire clinical trial process. For healthcare providers, tools like TrialGPT may streamline the identification of appropriate trials for patients, enhancing patient care by offering additional treatment options. For researchers, platforms like BEKplatform ensure a steady influx of qualified participants, reducing delays and increasing the likelihood of trial success.

Driving Increased Access and Participation

The combined use of AI technologies in clinical trial matching and recruitment addresses several longstanding challenges:

• Efficiency: Automating the matching process reduces the time required to identify suitable trials and participants, expediting the overall timeline of clinical research.
• Accuracy: AI algorithms can analyze complex datasets with a high degree of precision, ensuring that matches are based on comprehensive and up-to-date information.
• Accessibility: By streamlining the matching process, AI opens up clinical trial opportunities to a broader patient population, including those who may not have been aware of or had access to such options previously.
• Resource Optimization: Reducing the manual workload associated with matching allows clinicians and researchers to allocate their time and resources more effectively, focusing on patient care and study integrity.

AI-powered solutions like TrialGPT and BEKplatform are part of a significant shift in how patients make their way into clinical trials. By enhancing the capabilities of both healthcare providers and clinical researchers, these technologies facilitate a more efficient, accurate, and inclusive approach to patient recruitment and enrollment. The dual application of AI in matching patients to trials and identifying participants for studies holds the promise of accelerating medical research and improving patient outcomes, ultimately driving more eligible patients into studies and increasing access to clinical trials as a viable care option.