A Review on AI-Based Data-Driven Models for Optimization of Nanocarriers as Drug Delivery Systems

Artificial intelligence (AI), primarily machine learning (ML), has revolutionized as a transformative strategy to accelerate nanomedicine research and optimization. The present review highlights the nanocarrier-integrated AI-based approaches to enhance the performance of drug delivery systems. Traditional nanosystems are designed via trial-and-error experiments, which are expensive and time-consuming. ML approaches requiring distinct supervised and unsupervised algorithms and computational models offer a strong alternative. These models are well oriented to optimize the nanoparticle (NP) as the carrier and reveal its properties and complex interactions in the biological environment. Multiscale machine-learned modeling infrastructure (MuMMI), agent-based modeling (ABM), quantitative structure-activity relationship (QSAR), physiologically based pharmacokinetic (PBPK), and pharmacokinetic/pharmacodynamic (PK/PD) models are the varied models to predict NP synthesis parameters, nano-bio interactions, biodistribution, and nanotoxicity. These models improve the structure of the NP while minimizing the experimental burdens, elevating the prediction accuracy, and facilitating translational research. It also paves the way toward the next generation of smart and personalized medicine. Altogether, this review gives an overview of AI-driven techniques in nanomedicine, emphasizing their applications, benefits, and present obstacles.