Aplicaciones de inteligencia artificial en farmacovigilancia y detección de interacciones medicamentosas. Una revisión bibliográfica

Abstract

Abstract: Pharmacovigilance is key to ensuring drug safety during the post-marketing phase; however, conventional reporting methods have limitations when it comes to processing large volumes of data, which delays the detection of adverse reactions and drug interactions. To understand how to overcome this challenge, this study analyzes the impact of artificial intelligence (AI) in this field by evaluating its scope, limitations, ethical and regulatory challenges, and advantages. To this end, a structured narrative review was conducted using databases such as PubMed/MEDLINE, Scopus, and Web of Science, considering the SANRA recommendations and the PRIMA 2020 framework, resulting in the selection of 105 documents published between January 2018 and July 2026. The findings confirm that machine learning, deep learning, and natural language processing significantly streamline the detection of adverse event alerts and automate the analysis of reports; and although AI does not operate in isolation, it is an essential tool for ensuring clinical validation, algorithmic transparency, professional oversight, and compliance with current regulatory frameworks.