EUVIMEDEUVIMED entdecken
Crossref

Optimization of Nanoparticulate Drug Delivery Systems (NDDS) Using Hybrid Machine Learning Architectures: A Predictive Modeling Approach for Controlled Release Kinetics

Katta Vijay Kumar, A. V. Santhosh Kumar

International Journal of Drug Delivery Technology · 30.05.2026 · 16 (38s)

DOI: 10.25258/ijddt.16.38s.53 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 30.05.2026. Verfasst von Katta Vijay Kumar und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Nanoparticulate drug delivery systems (NDDS) offer superior therapeutic precision compared to conventional formulations; however, the experimental optimization of their controlled release kinetics remains resource-intensive and time-consuming. Conventional trial-and-error approaches are ill-suited for navigating the high-dimensional parameter space defined by polymer composition, particle morphology, drug loading, and environmental conditions.

Method

This study proposes a hybrid machine learning (ML) architecture that integrates Random Forest (RF) and Multilayer Perceptron (MLP) models to predict in-vitro drug release profiles of polymeric nanoparticles encapsulating Doxorubicin, Insulin, Paracetamol, and Cisplatin. A dataset comprising 1,200 simulated and experimentally validated formulation records was constructed, featuring nine physicochemical and process-related descriptors. The ensemble model was trained using stratified 5-fold cross-validation and evaluated against five baseline algorithms. Key Results: The proposed Hybrid RF+MLP architecture achieved an R² of 0.963, a root-mean-square error (RMSE) of 4.38%, and a prediction accuracy of 96.30%, outperforming standalone SVM (87.10%), KNN (84.30%), and MLP (94.10%) models. Bioavailability simulations demonstrated that the ML-predicted NDDS formulation maintained plasma drug concentrations above the minimum effective concentration (MEC) for 18 hours.

Conclusion

The proposed computational framework substantially reduces formulation development cycles, enabling data-driven design of controlled release nanomedicines with applicability to personalized oncology and chronic disease management.

Zitieren

APA

Kumar, K. V., & Kumar, A. V. S. (2026). Optimization of Nanoparticulate Drug Delivery Systems (NDDS) Using Hybrid Machine Learning Architectures: A Predictive Modeling Approach for Controlled Release Kinetics. International Journal of Drug Delivery Technology, 16(38s). https://doi.org/10.25258/ijddt.16.38s.53

Vancouver

Kumar KV, Kumar AVS. Optimization of Nanoparticulate Drug Delivery Systems (NDDS) Using Hybrid Machine Learning Architectures: A Predictive Modeling Approach for Controlled Release Kinetics. International Journal of Drug Delivery Technology. 2026;16(38s). doi:10.25258/ijddt.16.38s.53

BibTeX
@article{Kumar2026,
  author = {Kumar, Katta Vijay and Kumar, A. V. Santhosh},
  title = {Optimization of Nanoparticulate Drug Delivery Systems (NDDS) Using Hybrid Machine Learning Architectures: A Predictive Modeling Approach for Controlled Release Kinetics},
  journal = {International Journal of Drug Delivery Technology},
  year = {2026},
  volume = {16},
  number = {38s},
  doi = {10.25258/ijddt.16.38s.53}
}
Kostenlos starten →