ARTIFICIAL INTELLIGENCE APPLIED TO CLINICAL MEDICINE: POTENTIAL FOR RISK PREDICTION AND PERSONALIZATION OF CARE

Abstract

Introduction: Artificial intelligence (AI) applied to clinical medicine has established itself as a promising tool for enhancing the diagnostic, prognostic, and therapeutic capabilities of healthcare professionals. Machine learning and deep learning models have demonstrated the ability to identify complex patterns within vast amounts of clinical, laboratory, and imaging data, facilitating the prediction of adverse events, risk stratification, and early disease detection. Objective: To evaluate, through a systematic literature review, the scientific evidence regarding the application of AI in clinical medicine, with an emphasis on risk prediction, clinical decision support, early disease detection, and the personalization of care strategies. Methodology: The review was structured according to PRISMA checklist recommendations, utilizing scientific articles published over the last ten years and retrieved from the PubMed, SciELO, and Web of Science databases. The descriptors "Artificial Intelligence," "Machine Learning," "Clinical Medicine," "Risk Prediction," and "Personalized Medicine" were used, combined via Boolean operators. Included studies were those published in scientific journals, research related to the clinical application of AI, and works addressing risk prediction or the personalization of care. Duplicate studies, works unrelated to clinical practice, and publications lacking results relevant to the topic were excluded. Results: The studies demonstrated that AI has the potential to improve cardiovascular risk stratification, the prediction of mortality and hospitalization, the early identification of clinical events, and diagnostic support. Algorithms were also utilized to analyze patterns associated with diabetes, cardiovascular diseases, cancer, and other chronic conditions. Conclusion: It was concluded that artificial intelligence holds significant potential to transform clinical medicine, particularly by facilitating risk prediction, early diagnosis, and the individualization of care. Its use had expanded the capacity to analyze clinical information, but its safe incorporation still depended on scientific validation, transparency, data quality, and medical oversight.