A Literature Review of Artificial Intelligence in Health Care

Abstract

. Artificial intelligence (AI) is reshaping health care through medical-image interpretation, risk prediction, clinical decision support, natural-language processing, remote monitoring, and biomedical discovery. This article reviews the principal applications, benefits, limitations, and governance requirements of AI in health care. A systematic-style review protocol was designed around peer-reviewed and authoritative literature published from 2017 to 2025. To demonstrate an empirical journal format without misrepresenting a completed study, database retrieval, screening counts, coded outcomes, tables, and graphs are explicitly simulated. The illustrative synthesis contains 72 studies: medical imaging was the largest domain (30.6%), followed by predictive analytics using electronic health records (19.4%), clinical decision support (13.9%), and natural-language processing or large language models (12.5%). The most frequently coded benefit was improved diagnostic or classification performance, whereas limited generalizability, privacy risk, insufficient explainability, and workflow integration were the dominant barriers. The literature indicates that technical accuracy alone is insufficient for safe clinical value; external validation, representative data, human oversight, interoperability, regulatory evaluation, and continuous post-deployment monitoring are necessary. The review concludes that AI is most credible as an augmentative technology embedded within accountable clinical systems rather than as an autonomous replacement for health professionals.