How AI and Machine Learning Stand for Used in Keeping an Eye on Drug Safety a Review Article

Pharmacovigilance, as explained by the World Health Organization (WHO), involves various activities aimed at spotting and preventing any drug-related issues. Because there are many drug safety incidents, pharmaceutical companies and government health agencies believe that these activities are crucial for keeping patients safe. One major goal is to quickly find adverse drug events (ADEs), which are harmful effects that happen when a patient takes a medicine and might be caused by the drug. Artificial intelligence, using machine learning, employs algorithms and past knowledge to make predictions. Lately, there has been a growing interest in using more artificial intelligence in monitoring the safety of medicines already on the market and those in development. This study aimed to uncover and explain how artificial intelligence is used in pharmacovigilance by reviewing existing literature. We conducted a detailed analysis to compare the pros and cons of machine learning and deep learning, especially in tasks like identifying entities and classifying relationships related to ADE extraction. Furthermore, we looked at specific features and how they affect the performance of these methods. Broadly, our research also explored extracting ADEs from different sources like scientific papers, social media, and drug labels, not just relying on machine learning or deep learning alone.