The evolution and contemporary applications of artificial intelligence in medicine: a narrative review

Background and Objective: Artificial intelligence (AI) has rapidly emerged as a transformative technology in healthcare, driven by advances in computational power and the availability of large-scale medical datasets. AI techniques, including machine learning (ML), deep learning (DL), and generative AI, have demonstrated growing potential to improve clinical decision-making and support precision medicine. This narrative review aims to provide internal medicine and general physicians with an overview of AI and a contemporary understanding of its current and potential applications in medicine.

Methods: We conducted a narrative review using PubMed/MEDLINE, Scopus, and Web of Science, including English-language studies published up to March 31, 2026. The review focused on the evolution of AI technologies and their clinical applications across major areas of healthcare. Relevant studies and reviews describing ML, DL, and generative AI applications in diagnosis, treatment optimization, prognosis prediction, and patient engagement were evaluated. Key challenges related to AI implementation in clinical practice were also examined.

Key Content and Findings: AI has demonstrated promising capabilities in multiple domains of medicine. ML and DL models can analyze complex datasets, including medical imaging, laboratory results, and electronic health records, to enhance diagnostic accuracy and risk prediction. In addition, AI-driven approaches enable more personalized therapeutic strategies through precision medicine and multi-omic data integration. Generative AI models further expand these capabilities by supporting clinical documentation, decision support, and patient engagement tools. Despite these advances, important challenges remain, including limited model interpretability, algorithmic bias, patient privacy concerns, and barriers to clinical implementation.

Conclusions: AI holds substantial promise for improving healthcare delivery and patient outcomes. Addressing current limitations while ensuring responsible implementation will be essential for the safe and effective integration of AI into clinical practice.