ARTIFICIAL INTELLIGENCE IN HEALTH: FUNDAMENTALS, APPLICATIONS AND CHALLENGES

Abstract

Artificial intelligence (AI) has become a transformative tool in healthcare. Its main purpose is to assist, rather than replace, clinical judgment in tasks such as diagnosis, risk prediction, and personalized medicine. This work reviews the foundations of AI, its main training modalities (supervised, unsupervised, and reinforcement learning), and examples of medical applications including diabetes prediction from clinical data, sepsis phenotyping, and automated tumor localization in medical images. The main challenges identified include ethical and legal safe guards for patient data, model interpretability, and the prevention of biases that may reproduce healthcare inequities. Active participation of physicians and healthcare professionals in the design, supervision, and regulation of AI systems is essential to ensure that these technologies incorporate human values and clinical criteria. In conclusion, the balanced integration of artificial intelligence and human judgement will foster more precise, ethical, and patient-centered care.