Application of AI generated text-to-video in medical education: Systematic review

BackgroundTraditional medical education often struggles to simplify complex concepts for both healthcare professionals and patients. AI-generated text-to-video technologies are emerging as tools to enhance medical education by transforming intricate medical content into accessible visual formats. This systematic review aims to evaluate the current literature on the application of AI-generated text-to-video technologies in medical education. MethodsA comprehensive search was conducted in MEDLINE/PubMed, Google Scholar, Scopus, Cochrane Review, and Web of Science for studies published up to January 2025. The search targeted AI-generated text-to-video applications in medical education and patient engagement. Studies were screened based on predefined inclusion and exclusion criteria, and data were extracted independently. The risk of bias was assessed using the QUADAS-2 tool, and the review adhered to PRISMA guidelines. ResultsOut of 103 identified studies, 5 met the inclusion criteria. Four studies focused on patient education, and one on physician training. Applications spanned various specialties, including ophthalmology, neurosurgery, plastic surgery, and stroke rehabilitation. AI-generated videos showed potential to improve patient understanding, engagement, and confidence. However, limitations included data biases, content inaccuracies, lack of comparison with traditional methods, and variability in user technological proficiency. ConclusionAI-generated text-to-video technology holds promise for advancing medical education by improving engagement, enhancing learning outcomes, and facilitating patient understanding. Nevertheless, challenges related to data accuracy, algorithmic bias, ethical concerns, and equitable access must be addressed. Ongoing research, validation studies, and ethical oversight are essential to ensure the safe, effective, and inclusive integration of this technology in medical education.