Large language models in medicine: A systematic review of applications in medical, healthcare, and educational contexts

Large language models have emerged as transformative tools in medicine and medical education, offering applications in aided diagnosis, automation of clinical assessments, and optimization of healthcare workflows. This article critically reviews 112 relevant publications analyzing the use of LLMs in these fields. It explores their applications in specific tasks such as biomedical classification, automated clinical assessment, medical question answering, medical report generation, and enhancement in medical education through exam simulation and personalized tutoring. Despite their advances, LLMs continue to face significant challenges, including data privacy issues, clinical validation, and algorithmic biases. However, their integration into clinical and educational settings demonstrates considerable potential to improve efficiency, accuracy, and accessibility in health care, provided these models adhere to technical and ethical rigor. This article offers a comprehensive overview for healthcare professionals and researchers who aim to adopt these models responsibly.