Artificial Intelligence in Healthcare: Clinical Applications, Decision-Making, and Implementation Challenges

Objective: Artificial intelligence (AI) is increasingly influencing healthcare, yet its impact is uneven across clinical settings. Some applications already support diagnosis and workflow, while others remain in earlier developmental stages. This review examined the main clinical applications of AI and the barriers limiting broader adoption.Methods: We performed a structured search of PubMed, Scopus, Embase, and Web of Science for articles published between January ۲۰۱۰ and April ۲۰۲۴. Search terms covered artificial intelligence, machine learning, healthcare, medical imaging, predictive analytics, robotics, and virtual assistants. Studies addressing clinically relevant AI applications were included; editorials, conference abstracts, non-healthcare studies, and articles lacking full text were excluded.Results: ۴۱ studies met the inclusion criteria. The most consistent progress appeared in imaging, predictive analytics, and clinical workflow support. In imaging, AI improved detection, classification, and segmentation, though performance proved more reliable on curated datasets than in routine practice. Predictive models were commonly used for risk stratification, disease prediction, and outcome estimation, often leveraging electronic health record data. Workflow-oriented tools, such as documentation support and triage systems, integrated more readily and delivered efficiency gains. Robotics and assistive technologies varied in maturity, constrained by cost, infrastructure, and training demands.Conclusions: AI contributes to several areas of healthcare, yet its value depends heavily on implementation. Data quality, bias, limited interpretability, interoperability, and regulatory uncertainty remain major barriers to broader use. Rather than replacing clinical expertise, AI functions best as a supportive tool. Its value emerges when validated in real-world settings, aligned with clinical needs, and governed within a clear framework.

کلیدواژه ها:

Artificial Intelligence ، Healthcare Systems ، Clinical Decision-Making ، Predictive Analytics ، Medical Imaging ، AI Adoption