Evaluating the Efficacy of AI Chatbots as Tutors in Urology: A Comparative Analysis of Responses to the 2022 In-Service Assessment of the European Board of Urology.

INTRODUCTION: This study assessed the potential of large language models (LLMs) as educational tools by evaluating their accuracy in answering questions across urological subtopics. METHODS: Three LLMs (ChatGPT-3.5, ChatGPT-4, and Bing AI) were examined in two testing rounds, separated by 48 h, using 100 Multiple-Choice Questions (MCQs) from the 2022 European Board of Urology (EBU) In-Service Assessment (ISA), covering five different subtopics. The correct answer was defined as "formal accuracy" (FA) representing the designated single best answer (SBA) among four options. Alternative answers selected from LLMs, which may not necessarily be the SBA but are still deemed correct, were labeled as "extended accuracy" (EA). Their capacity to enhance the overall accuracy rate when combined with FA was examined. RESULTS: In two rounds of testing, the FA scores were achieved as follows: ChatGPT-3.5: 58% and 62%, ChatGPT-4: 63% and 77%, and BING AI: 81% and 73%. The incorporation of EA did not yield a significant enhancement in overall performance. The achieved gains for ChatGPT-3.5, ChatGPT-4, and BING AI were as a result 7% and 5%, 5% and 2%, and 3% and 1%, respectively (p > 0.3). Within urological subtopics, LLMs showcased best performance in Pediatrics/Congenital and comparatively less effectiveness in Functional/BPS/Incontinence. CONCLUSION: LLMs exhibit suboptimal urology knowledge and unsatisfactory proficiency for educational purposes. The overall accuracy did not significantly improve when combining EA to FA. The error rates remained high ranging from 16 to 35%. Proficiency levels vary substantially across subtopics. Further development of medicine-specific LLMs is required before integration into urological training programs.

How does artificial intelligence master urological board examinations? A comparative analysis of different Large Language Models' accuracy and reliability in the 2022 In-Service Assessment of the European Board of Urology.

PURPOSE: This study is a comparative analysis of three Large Language Models (LLMs) evaluating their rate of correct answers (RoCA) and the reliability of generated answers on a set of urological knowledge-based questions spanning different levels of complexity. METHODS: ChatGPT-3.5, ChatGPT-4, and Bing AI underwent two testing rounds, with a 48-h gap in between, using the 100 multiple-choice questions from the 2022 European Board of Urology (EBU) In-Service Assessment (ISA). For conflicting responses, an additional consensus round was conducted to establish conclusive answers. RoCA was compared across various question complexities. Ten weeks after the consensus round, a subsequent testing round was conducted to assess potential knowledge gain and improvement in RoCA, respectively. RESULTS: Over three testing rounds, ChatGPT-3.5 achieved RoCa scores of 58%, 62%, and 59%. In contrast, ChatGPT-4 achieved RoCA scores of 63%, 77%, and 77%, while Bing AI yielded scores of 81%, 73%, and 77%, respectively. Agreement rates between rounds 1 and 2 were 84% (κ = 0.67, p < 0.001) for ChatGPT-3.5, 74% (κ = 0.40, p < 0.001) for ChatGPT-4, and 76% (κ = 0.33, p < 0.001) for BING AI. In the consensus round, ChatGPT-4 and Bing AI significantly outperformed ChatGPT-3.5 (77% and 77% vs. 59%, both p = 0.010). All LLMs demonstrated decreasing RoCA scores with increasing question complexity (p < 0.001). In the fourth round, no significant improvement in RoCA was observed across all three LLMs. CONCLUSIONS: The performance of the tested LLMs in addressing urological specialist inquiries warrants further refinement. Moreover, the deficiency in response reliability contributes to existing challenges related to their current utility for educational purposes.