Mishra Hara Prasad, Gupta Rachna
Department of Pharmacology, University College of Medical Sciences, University of Delhi, Delhi, India.
Curr Rev Clin Exp Pharmacol. 2025;20(2):89-97. doi: 10.2174/0127724328311400240823062829.
Predictions are made by artificial intelligence, especially through machine learning, which uses algorithms and past knowledge. Notably, there has been an increase in interest in using artificial intelligence, particularly generative AI, in the pharmacovigilance of pharmaceuticals under development, as well as those already in the market. This review was conducted to understand how generative AI can play an important role in pharmacovigilance and improving drug safety monitoring. Data from previously published articles and news items were reviewed in order to obtain information. We used PubMed and Google Scholar as our search engines, and keywords (pharmacovigilance, artificial intelligence, machine learning, drug safety, and patient safety) were used. In toto, we reviewed 109 articles published till 31st January 2024, and the obtained information was interpreted, compiled, evaluated, and conclusions were reached. Generative AI has transformative potential in pharmacovigilance, showcasing benefits, such as enhanced adverse event detection, data-driven risk prediction, and optimized drug development. By making it easier to process and analyze big datasets, generative artificial intelligence has applications across a variety of disease states. Machine learning and automation in this field can streamline pharmacovigilance procedures and provide a more efficient way to assess safety-related data. Nevertheless, more investigation is required to determine how this optimization affects the caliber of safety analyses. In the near future, the increased utilization of artificial intelligence is anticipated, especially in predicting side effects and Adverse Drug Reactions (ADRs).
预测由人工智能做出，尤其是通过机器学习，机器学习利用算法和过往知识。值得注意的是，在研发中的药品以及已上市药品的药物警戒中，使用人工智能，特别是生成式人工智能的兴趣有所增加。进行本次综述是为了了解生成式人工智能如何在药物警戒和改善药物安全监测中发挥重要作用。我们查阅了先前发表的文章和新闻报道中的数据以获取信息。我们使用PubMed和谷歌学术作为搜索引擎，并使用了关键词（药物警戒、人工智能、机器学习、药物安全和患者安全）。总体而言，我们查阅了截至2024年1月31日发表的109篇文章，并对所获信息进行了解释、汇编、评估并得出结论。生成式人工智能在药物警戒方面具有变革潜力，展现出诸多益处，如增强不良事件检测、数据驱动的风险预测以及优化药物研发。通过使处理和分析大型数据集变得更加容易，生成式人工智能在各种疾病状态下都有应用。该领域的机器学习和自动化可以简化药物警戒程序，并提供一种更有效的方式来评估与安全相关的数据。然而，需要更多研究来确定这种优化如何影响安全分析的质量。在不久的将来，预计人工智能的使用将会增加，尤其是在预测副作用和药物不良反应（ADR）方面。
