International Journal of Drug Delivery Technology
Volume 16, Issue 2, 2026

Development of a Machine Learning Assisted Model for Formulating Modified Release Drug Delivery System

Shivani J. Gandhi1,2*, Punit B. Parejiya3, Shivangi J. Gandhi4, Shetal Desai5, Vidhi Patel5, Bijal Yadav5, Ankitkumar N. Patel6, Niravbhai J. Patel7

1PhD Research Scholar, Kadi Sarva Vishwavidyalaya, Gandhinagar, Gujarat, India

2Faculty of Pharmacy, The Maharaja Sayajirao University of Baroda, Baroda, Gujarat, India

3K.B. Institute of Pharmaceutical Education and Research, Kadi Sarva Vishwavidyalaya, Gandhinagar, Gujarat, India

4GLS University, Ahmedabad, Gujarat, India

5Smt. B.N.B Swaminarayan Pharmacy College, Salvav, Vapi, Gujarat, India

6Director, Formulation R&D, Amneal Pharmaceuticals.

7Vice President, R&D Nivagen Pharmaceuticals Inc.

*Corresponding Author: Ms. Shivani J. Gandhi, PhD Research Scholar, Kadi Sarva Vishwavidyalaya, Sector-15, Gandhinagar, Gujarat, India. Email: shivanigandhi246@gmail.com

ABSTRACT

Background: Modified-release medication delivery systems are crucial to achieve regulated therapeutic outcomes. This work develops a machine learning-enhanced prediction model for sustained-release tablets using hydrophilic (HPMC), hydrophobic (Eudragit), and composite polymers.

Methods: Such machine-learning techniques include the Support Vector Machine (SVM), Ridge regression (RR), Random Forest (RF), and Decision Tree (DT), which were employed to optimize the formulation parameters. The dataset was appropriately split into training, validation, and test sets. Evaluation criteria, including accuracy, root mean squared error (RMSE), and mean absolute error (MAE), were used to assess model performance. Its accuracy and generalization ability characterized the optimum model.

Conclusion: This study showed that machine learning models can predict drug release with confidence, aiding the formulation process for improved modified-release medication delivery.

Keywords: Modified-release drug delivery, machine learning, sustained-release tablets, HPMC, Eudragit, support vector machine, k-nearest neighbours, ridge regression, random forest, decision tree, drug formulation prediction.

How to cite this article: Gandhi SJ, Parejiya PB, Gandhi SJ, Desai S, Patel V, Yadav B, Patel AN, Patel NJ. Development of a Machine Learning Assisted Model for Formulating Modified Release Drug Delivery System. Int J Drug Deliv Technol. 2026;16(2): 693-698. DOI: 10.25258/ijddt.16.2.74

Source of support: Nil.

Conflict of interest: None