This note synthesises key developments in the use of artificial intelligence (AI) in healthcare over the 2014–2023 period, with emphasis on clinical applications and governance constraints. It reviews advances in machine learning for diagnosis and prognosis, predictive analytics for personalised medicine, natural language processing for clinical text mining, and computer vision for medical imaging. The discussion highlights the statistical and data science foundations required for robust development and evaluation, including external validation, calibration, subgroup assessment, and monitoring for dataset shift. Methodologically, the synthesis is informed by a structured search in the Web of Science Core Collection (conducted 21 May 2023), targeting English-language review articles (publication years 2020–2023) and complemented by foundational works widely cited in the field. Findings indicate that AI systems can improve diagnostic precision and operational efficiency when embedded within accountable clinical workflows. However, adoption is constrained by persistent challenges related to explainability, interoperability, bias, privacy protection, and institutional governance. The note concludes that AI should be treated as clinical augmentation under human oversight, with transparent governance and rigorous evaluation as prerequisites for sustainable deployment. Keywords: artificial intelligence in healthcare; machine learning; predictive analytics; natural language processing; medical imaging; health data governance.
Milena-Jael Silva-Morales (Sat,) studied this question.