What You Should Know:
– A new study published in the Journal of Medical Artificial Intelligence explores the potential of large language models (LLMs), like GPT-4, to act as valuable tools for clinicians.
– Study author Marc D. Succi, MD, strategic innovation leader at Mass General Brigham Innovation, and his team investigated the use of GPT-4 in recommending physical exam instructions based on patient symptoms. For example, if a patient presented with a painful hip, GPT-4 would generate a detailed list of physical examination steps for the clinician to follow.
Results/Outcomes
To assess the quality of these AI-generated instructions, three attending physicians evaluated GPT-4’s recommendations across various symptoms. They scored the LLM on accuracy, comprehensiveness, readability, and overall quality. Impressively, GPT-4 scored high marks, achieving at least 80% of the possible points across all symptoms.
Despite these limitations, the study’s findings suggest that LLMs hold significant promise for the future of healthcare. They could serve as valuable resources for clinicians, particularly those with less experience or those practicing in resource-limited settings. By providing comprehensive and accurate physical exam instructions, LLMs could help bridge knowledge gaps and improve diagnostic accuracy, ultimately leading to better patient outcomes.
“Medical professionals early in their career may face challenges in performing the appropriate patient-tailored physical exam because of their limited experience or other context-dependent factors, such as lower resourced settings,” said senior author Marc D. Succi, MD, strategic innovation leader at Mass General Brigham Innovation, associate chair of innovation and commercialization for enterprise radiology and executive director of the Medically Engineered Solutions in Healthcare (MESH) Incubator at Mass General Brigham. “LLMs have the potential to serve as a bridge and parallel support physicians and other medical professionals with physical exam techniques and enhance their diagnostic abilities at the point of care.”
