Use of Artificial Intelligence and Machine Learning in Rapid Drug Discovery and Pharmacovigilance
Ogboh Rita Onyebuchi
*
Faculty of Pharmacy, Niger Delta University, Bayelsa, Nigeria.
Olukunle O. Akanbi
Department of Psychology and Behavioral Sciences, National Louis University, FL, USA.
Adepeju Kafayat Olowookere
Department of Nanoscience, University of North Carolina at Greensboro, USA.
Isaac Baiden
Department of Physiology, Wayne State University, Detroit, MI 48202, United States.
Ngwoke, Faith Chinaza
School of Pharmacy and Health Sciences, United States International University Africa, Nairobi, Kenya.
Onodje Lucky
Department of Computing and Mathematics, Manchester Metropolitan University, United Kingdom.
Akinlolu Oluwamisimi
Johns Hopkins University, Carey Business School, Maryland, USA.
Ndidi Atasie Eboh
Department of Nursing, Stark State College North Canton OH, United States.
Uchechukwu Lilian Okoye
Department of Chemistry, Georgia Southern University, Georgia, USA.
*Author to whom correspondence should be addressed.
Abstract
The increasing global burden of disease, rising research and development costs, and high attrition rates in pharmaceutical pipelines underscore the need for more efficient approaches to therapeutic development and drug safety monitoring. Artificial intelligence (AI) and machine learning (ML) have emerged as data-driven tools with the potential to improve multiple stages of the pharmaceutical lifecycle.
This narrative review is designed to provide a structured and critical overview of the use of AI and ML in drug discovery and drug safety surveillance. A comprehensive literature search was performed to identify relevant studies using major electronic databases, with emphasis on publications from 1997 to March 2025. The studies were selected based on the inclusion criteria.
The findings of the study show that AI and ML are being used in drug discovery, drug development, and drug safety surveillance. These technologies have the potential to provide predictive models, integrate heterogeneous biomedical data, and analyze real-world data to detect adverse drug reactions. Deep learning and natural language processing have been found to be useful tools to improve early risk detection.
However, some limitations have also been found. These include the quality of the data, bias in AI and ML models, lack of interpretability of AI and ML models, lack of external validation, and lack of real-world implementation. These limitations need to be addressed to make AI and ML more useful tools for drug discovery and drug safety surveillance.
Overall, while AI and ML offer meaningful opportunities to enhance drug discovery and pharmacovigilance, their impact remains dependent on rigorous validation, improved data governance, and alignment with clinical and regulatory frameworks. Continued research and context-specific implementation strategies will be essential to support their effective and equitable integration into pharmaceutical research and healthcare systems.
Keywords: Artificial intelligence, machine learning, drug discovery, pharmacovigilance, adverse drug reactions.