Machine learning empowered formulation design, optimization and characterization of nanoparticulate drug delivery systems: Current applications, challenges, and future perspectives
Chunyan Shen, Mengyan Zhang, Meiting Lu, Errong Chang, Ziting Gao, Weikang Ban, Qiang Liu, Zhong Zuo, Cuiping Jiang
Abstract
Nanoparticulate drug delivery systems (NDDS) have revolutionized modern medicine by significantly improving drug targeting, bioavailability, and therapeutic efficacy. Despite the clinical success of over 90 approved nanomedicines, the development of NDDS remains challenging due to the complexity of formulation design, optimization, and characterization processes. Artificial intelligence, particularly machine learning (ML), offers powerful data analytics and predictive capabilities that can address these challenges. This review systematically summarizes recent advances in ML applications across various NDDS formulations, including polymeric nanoparticles, lipid nanoparticles, liposomes, solid lipid nanoparticles, nanostructured lipid carriers, nanoemulsions, nanosuspensions, lipid-based hybrid NDDS, self-emulsifying drug delivery systems, niosomes, and nanocrystals. We also summarize how ML algorithms could help predict critical quality attributes of NDDS, such as particle size, shape, surface properties, drug encapsulation efficiency, drug loading efficiency, drug release behavior, and stability. Furthermore, we discuss existing challenges and prospects for the formulation development empowered by ML in NDDS. In conclusion, this review provides a comprehensive overview of the transformative potential of ML in improving the formulation development of nanomedicines, ultimately accelerating their clinical translation.