Artificial Intelligence Driven Approaches to Predicting Drug Toxicity: Challenges and Future Direction | ||
| Egyptian Society of Clinical Toxicology Journal | ||
| Articles in Press, Accepted Manuscript, Available Online from 17 November 2025 | ||
| Document Type: Review Article | ||
| DOI: 10.21608/esctj.2025.411832.1098 | ||
| Author | ||
| Shrouk Mohamed Ali* | ||
| Forensic Medicine and Toxicology, Faculty of Medicine, Suez Canal University | ||
| Abstract | ||
| Background: Drug-induced toxicity is major challenges in drug discovery, threatening patient safety and dramatically increasing healthcare expenditures. The integration of Artificial Intelligence (AI) into toxicological research is reshaping the landscape of prediction of drug toxicity, offering significant advancements in safety assessment and reducing reliance on animal testing. Machine learning (ML) and deep learning (DL) present a powerful alternative by enabling the analysis of large-scale datasets from preclinical and clinical sources to uncover complex toxicological patterns. However, there are several critical challenges persist in applying. Objectives: To explore the current landscape of AI-driven toxicological research and future directions, opportunities, and challenges associated with implementing AI in drug toxicity prediction. Methods: a comprehensive theoretical examination of the role of (AI) and (ML) in the prediction of drug toxicity provides in this review by summarizes global literature. It covers methodologies ranging from data mining to deep learning, along with key databases, modeling algorithms, and software used for toxicity prediction. Furthermore, the progress of AI technologies particularly ML and deep learning models is discussed, highlighting their advantages, challenges, and potential future directions. Conclusion and Recommendations: AI-based approaches hold great promise for drug toxicity prediction, but their real-world reliability is limited by challenges in model interpretability and translational validation. Overcoming these barriers demands interdisciplinary collaboration and responsible integration of AI into drug development alignment with implementation of regulatory and ethical safeguards. Together, these efforts will enhance reliability, facilitate translational applications, and maximize the real-world impact of AI in drug toxicity prediction. | ||
| Keywords | ||
| Artificial intelligence; Machine Learning; QSAR; Deep learning; Drug Toxicity | ||
|
Statistics Article View: 1 |
||
