Artificial Intelligence: A Current and Updated Review of Repurposed Drugs for SARS-CoV-2

Introduction: Expert systems are specific applications of artificial intelligence (AI) in pharmacy practice. This review explores various AI applications in optimizing pharmacotherapy using repurposed drugs for COVID-19 caused by SARS-CoV-2. Methods: This is a focused literature review with keywords relevant to the terms used in PubMed, Scopus, and Web of Science. A total of 150 in-depth, relevant, and high-quality studies were reviewed in this study. Results: Drug repositioning, reprofiling, or re-tasking could be used as an approach for recognizing new targets for accepted or tentative medications that were not initially approved or designated for COVID-19. The use of AI to optimize repurposed drugs for COVID-19 is a rapidly evolving and dynamic field, with new approaches and findings continually emerging. Future diagnosis and management could be expedited by AI assistance based on healthcare records and genetic data. In this direction, deep multi-layer recurrent neural networks, Random Forest (which combines the output of multiple decision trees to reach a single result), and the optimized distributed gradient boosting library (XGBoost) enable experienced medical workers to make predictions based on data. Discussion: Methods based on AI could increase pharmacotherapy efficiency in different diseases. In the preliminary study, several repurposed drugs, such as ritonavir, lopinavir, ivermectin, remdesivir, and others, have emerged as effective treatment strategies. Conclusion: Limitations in AI artifact development, precise instruction provisions, challenges in confirming interpretability, and concerns over time are obstacles to reliable AI-based outcomes for repurposed drugs in SARS-CoV-2 patients. Targeting multiple sites may be more effective because the mutability of RNA viruses can lead to drug resistance.