Mapping Personalized Learning in Medical Education: A Meta-Synthesis of Artificial Intelligence Applications
Introduction: The rapid advancement of Artificial Intelligence (AI) in medical education is driving a shift from traditional instructional design methods toward personalized and adaptive learning models. Despite numerous promising applications, the available evidence remains limited and fragmented; therefore, a comprehensive synthesis of the evidence is needed to support robust conclusions. Methods: This study employed the four-phase meta-synthesis framework proposed by Sandelowski and Barroso. A systematic search was conducted across Medline, Embase, CINAHL, PsycINFO, PubMed, Web of Science, ScienceDirect, Wiley Online Library, SpringerLink, Taylor & Francis Online, SAGE Journals, and Scopus, covering publications from 2010 to 2025. Studies were screened according to predefined inclusion and exclusion criteria, and their methodological quality was evaluated using the Critical Appraisal Skills Programme (CASP). Coding reliability was assessed through a test–retest procedure, resulting in a reliability coefficient of 0.81. Results: A total of 273 records were identified, of which 16 studies met the inclusion criteria and obtained CASP scores exceeding the threshold of 30. Content analysis revealed five principal domains: faculty-related applications (21%), student-related applications (28%), applications in the learning process (15%), curriculum development (13%), and assessment mechanisms (23%). Student-related applications constituted the largest proportion, highlighting the pivotal role of learner-centered personalization in AI-driven medical education. Conclusion: The integration of Artificial Intelligence (AI) into individualized educational experiences represents a transformative model for medical education. AI enables adaptive learning pathways, dynamic assessment methods, and data-driven instructional environments, thereby enhancing student engagement, fostering faculty innovation, and promoting equity in learning outcomes. This synthesis proposes an overarching conceptual framework to inform policy, research, and implementation in the context of AI-supported personalized medical education.