Aprendizagem baseada em problemas na tutoria do ensino de medicina: integrando o ChatGPT como ferramenta de suporte

Abstract

Introduction: Problem-Based Learning (PBL) in medical education tutoring, integrating ChatGPT as a support tool, represents a significant evolution in the educational field. It enables personalized learning experiences, more precise feedback, and adapts content to individual student needs. Objective: To explore how AI can be effectively integrated into the educational context, aiming to enhance students' learning experiences. Methodology: The systematic mapping methodology is a systematic review approach designed to map primary and secondary research articles to provide a comprehensive overview of a specific field of study. This methodology is particularly useful for identifying patterns, trends, research gaps, and areas of greater or lesser exploration in a given knowledge domain. Unlike traditional systematic reviews, which aim to answer specific questions, systematic mapping focuses on categorizing and classifying existing studies, offering a broad overview of the current research landscape. In this study, a qualitative, systematic, and descriptive mapping methodology was employed to identify the effectiveness of ChatGPT in medical education. Inclusion criteria, keywords, and search terms were defined to select articles for mapping. The chosen descriptors were: ChatGPT (ChatGPT), Teaching, Medicine, Learning, Artificial Intelligence, "Evaluation Studies as Topic," "Educational Measurement," and "Education, Medical, Undergraduate." Searches were conducted using search strings and Boolean operators (AND, OR) in virtual libraries such as PubMed, CAPES Journals, Wiley, BVS, SciELO, and Science Direct, crossing descriptors in Portuguese and English. The search strings used were: Portuguese: ChatGPT AND medicina AND (ensino OR aprendizagem OR avaliação) end English: ChatGPT AND medicine AND (teaching OR learning OR assessment). After applying the search strings, the articles were evaluated and classified according to the established inclusion criteria. This selection identified emerging thematic categories related to the use of ChatGPT in medical education. Based on the collected evidence, a systematization was conducted to highlight the main trends and approaches regarding ChatGPT’s effectiveness as a pedagogical tool in medical education. The generated work provides a step-by-step guide for exploring innovative ways to integrate digital technologies into education, aiming to improve pedagogical practices in the medical context. Thus, the research contributes scientific evidence that can guide future investigations and educational practices, helping to overcome challenges in teaching and promoting innovation in the use of educational technologies. Results and Discussion: The analysis of 50 selected articles and transcriptions of interactions with ChatGPT version 3.5, based on Bardin's analysis, revealed that integrating this technology into Problem-Based Learning (PBL) tutoring in medical education brought significant advancements. Adapting to the tool was not without challenges but demonstrated the resilience of educators and their ability to reinvent pedagogical practices. ChatGPT made teaching more dynamic and collaborative, fostering greater student autonomy and facilitating the resolution of complex clinical cases. Tutoring evolved into an interactive space where technology and the humanization of education coexist productively, highlighting the transformative potential of PBL in medical education. The tool's integration not only overcame difficulties but also opened new perspectives for the future of medical education. These studies formed the basis for creating the e-book PROBLEM-BASED LEARNING IN MEDICAL EDUCATION TUTORING: Integrating ChatGPT as a Support Tool.Conclusion: Integrating ChatGPT as a support tool in Problem-Based Learning (PBL) in medical education has proven to be a promising strategy to enrich the educational process, offering personalized, interactive, and effective learning experiences. The study analysis indicated that despite its great potential to enhance teaching, ChatGPT faces challenges related to accuracy, reliability, and critical reasoning ability, requiring constant supervision. When used carefully and ethically, ChatGPT can be a valuable complement to medical education, promoting greater student autonomy and facilitating the resolution of complex clinical cases. However, its effectiveness depends on balanced integration, respecting human expertise and educational supervision. Continued research and technological evolution are essential to maximize the benefits of ChatGPT in medical education, ensuring it serves as an ally in forming competent and ethical healthcare professionals.