Research Mentor Name
Eliza Beal
Research Mentor Email Address
beale@karmanos.org
Institution / Department
Wayne State University, Karmanos Cancer Institute
Document Type
Research Abstract
Research Type
healthcommunityimpact
Graduate Level Research
no
Abstract
Background:
Incidence and prevalence of Pancreatic Neuroendocrine Tumors (PNETs) has increased in recent years. More than half of American adults endorse utilizing the internet within the past year to search for medical information. This study aims to investigate the use of Large Language Models (LLMs), a popular type of artificial intelligence, to improve the efficacy of online patient education materials (oPEMs) in providing accurate and digestible information to patients with a PNET diagnosis.
Methods: Content from 36 web pages related to PNET was transferred to two popular LLMs, Chat GPT and Google Gemini, with the prompt “rewrite the following text at a sixth grade reading level”. Data was then collected on quality, readability and comprehensiveness of the revised oPEMs and compared to scores for the original text.
Results: For oPEMs with a mean Flesch-Kincaid Grade Level (FKGL) of 10, Chat GPT and Google Gemini created a mean grade level reduction of 2.30 and 2.86, respectively. Additionally, the majority of oPEM scores for quality and comprehensiveness were not significantly reduced by LLM editing. Gemini performed significantly better than GPT in FKGL reduction and quality maintenance (both p< 0.01), while no significant difference was observed between the two in maintaining comprehensiveness.
Conclusions: The use of LLMs resulted in a significant decrease in reading level of oPEMs, while mostly maintaining quality and comprehensiveness. Implementing LLMs, particularly Google Gemini, in this way can help improve patient understanding of materials regarding PNETs.
Disciplines
Medicine and Health Sciences | Neoplasms | Oncology | Surgery
Recommended Citation
Dolan, Brendan; Crayton, Aniela; Woodbury, Emerson; Tzita, Miguel; Beal, Eliza MD; and Dyson, Gregory PhD, "The Use of Large Language Models in Enhancing Patient Education Materials for Pancreatic Neuroendocrine Tumors" (2026). Medical Student Research Symposium. 475.
https://digitalcommons.wayne.edu/som_srs/475