From trial-and-error to inverse design: how AI is redefining drug delivery systems
Abstract
AI is transforming drug delivery system (DDS) design by enabling predictive modeling and inverse design strategies. Recent advances integrating machine learning, deep learning and physics-informed approaches have improved the ability to capture complex formulation-performance relationships. This review provides a structured overview of AI methodologies in DDS, compares their strengths and limitations, and discusses their integration into pharmaceutical pipelines. Key challenges, including data heterogeneity, limited generalizability, model interpretability and regulatory validation, are critically assessed as barriers to clinical translation of AI-driven DDS design.
Fields of Study & Subjects
Computational Drug Discovery Methods
Drug Solubulity and Delivery Systems
Machine Learning in Materials Science
Publication details
- Journal / source
- Drug Discovery Today
- Publisher
- Elsevier BV
- Volume
- 31
- Issue
- 4
- Pages
- 104725–104725
- Publication date
- 18 June 2026
- Language
- EN