Transformation in Medical Education with Generative AI: From Advanced Generative Models to Enhanced Clinical Skills

Background and Aims: With the rise of digital technologies, generative artificial intelligence (Generative AI), through advanced models such as GPT-۴, Med-PaLM, and Stable Diffusion, has introduced transformative possibilities for medical education. These tools enable the generation of precise educational content and realistic clinical simulations, addressing limitations of traditional teaching methods. This study aimed to assess the impact of generative AI on improving diagnostic accuracy, accelerating clinical decision-making, and enhancing learning satisfaction among medical students and specialized residents. Methods: A randomized controlled trial was conducted at a teaching hospital and medical university from January to June ۲۰۲۴. A total of ۱۵۰ medical students and residents were randomly assigned to three equal groups (n=۵۰ each): Control Group: traditional in-person teaching with standard materials. Experimental Group ۱: AI-based education using simulated clinical cases developed by Med-PaLM and ChatGPT. Experimental Group ۲: blended learning combining traditional and generative AI methods. Evaluations included diagnostic accuracy (via image- and case-based tests), response times in simulated clinical scenarios, and learning satisfaction (via standardized questionnaires). Assessments occurred at three points: before training, immediately after, and one month post-training. Data analysis involved one-way ANOVA and multivariate regression. Results: Average diagnostic accuracy scores were ۷۲% (Control), ۸۵% (Experimental ۱), and ۹۱% (Experimental ۲), with significant differences across groups (p < ۰.۰۱). The hybrid group (Experimental ۲) showed a ۴۰% reduction in decision-making time and a ۹۳% satisfaction rate, compared to ۷۵% in the control group. These findings highlight the effectiveness of generative AI in advancing both clinical reasoning and learner engagement. Conclusion: Generative AI, when integrated with traditional educational methods, can significantly enhance clinical training outcomes. Its ability to provide personalized, interactive, and realistic learning experiences marks it as a valuable tool for modern medical education. Future studies should explore long-term impacts, develop implementation frameworks, and address ethical considerations surrounding AI use in education.

کلیدواژه ها:

GenerativeAI ، MedicalEducation ، ClinicalSimulation ، Artificial Intelligence ، Advanced Generative Models