Comparative Analysis of the Readability of Artificial Intelligence-Supported Educational Materials for Acute Cholecystitis Patients: ChatGPT and Gemini
Introduction: Acute cholecystitis is a common gastrointestinal disease characterized by inflammation of the gallbladder, and patient education constitutes an important component of the treatment process. In recent years, artificial intelligence (AI)-based language models have increasingly been used in the preparation of patient education materials. However, the readability level of these materials is critically important for patients to understand health information. This study aimed to compare the readability of acute cholecystitis educational materials generated by ChatGPT and Gemini.
Methods: In this comparative study, AI-generated patient education texts about acute cholecystitis were analyzed. Twenty texts were obtained from OpenAI ChatGPT and twenty from Gemini. Readability was evaluated using the Average Reading Level Consensus, Automated Readability Index (ARI), Flesch Reading Ease, Fog Scale, Flesch–Kincaid Grade Level, Coleman–Liau Index, SMOG Index, Linsear Write Formula, and Forcast Readability Formula. Differences between groups were analyzed statistically, and p<0.05 was considered significant.
Results: Texts generated by Gemini demonstrated higher readability in the Flesch Reading Ease score (58.48±3.42 vs. 44.28±4.36; p<0.001). Fog Scale results also revealed that Gemini content was easier to read (11.44±1.42 vs. 15.48±1.36; p<0.001). In terms of the Automated Readability Index, ChatGPT showed lower scores (9.32±2.38 vs. 12.42±2.44; p=0.032). No significant differences were found in other readability indices.
Conclusion: Both AI models produced readable texts that could be used for patient education regarding acute cholecystitis. However, Gemini provided more understandable content in some important readability measures. Although AI-supported patient education materials have the potential to improve health literacy, readability should be optimized for broader patient populations.
Keywords: Acute cholecystitis, Artificial intelligence, Readability, ChatGPT, Gemini
