A computational framework for data-driven drug delivery design: computational validation via drug release kinetics prediction

Accurate prediction of drug release plays an important role in the design of controlled-release pharmaceutical formulations. Traditional formulation development relies heavily on experimental trial-and-error approaches, which are often time-consuming and resource-intensive. This study presents a computational framework for predicting cumulative drug release from key formulation variables and evaluates its performance using a simulated dataset generated with the Korsmeyer-Peppas model. The dataset consisted of 180 simulated formulation samples described by six input variables: polymer concentration, drug load, particle size, dissolution medium pH, dissolution time, and viscosity. Two supervised learning models, Random Forest (RF) and an Artificial Neural Network (ANN), were developed and evaluated using an 80:20 train-test split and five-fold cross-validation. The ANN achieved higher predictive performance than the RF model, with a test R2 of 0.9652, RMSE of 6.855%, and MAE of 6.060%, while the RF model achieved a test R2 of 0.5901, RMSE of 23.539%, and MAE of 16.437%. Feature importance analysis identified particle size and polymer concentration as the most influential variables affecting the predicted drug release profiles. The findings show that machine learning can effectively model relationships between formulation variables and simulated drug release behaviour, and provide a reproducible framework for computational evaluation during the early stages of formulation development. The reported results are based on simulated data and should be interpreted as computational validation of the proposed framework. Future studies will focus on evaluating the framework using experimentally measured dissolution datasets.

Authors

Institutions

Publication Details

Journal
Frontiers in Drug Delivery
Published
2026-11-19
DOI
https://doi.org/10.3389/fddev.2026.1941896
Primary Topic
Advanced Drug Delivery Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A computational framework for data-driven drug delivery design: computational validation via drug release kinetics prediction

Anthony Amaechi Attama, Adaeze Linda Onugwu, Chinazom Precious Agbo, Franklin Chimaobi Kenechukwu et al.
Frontiers in Drug Delivery
Advanced Drug Delivery Systems
article

A computational framework for data-driven drug delivery design: computational validation via drug release kinetics prediction

Anthony Amaechi Attama, Adaeze Linda Onugwu, Chinazom Precious Agbo, Franklin Chimaobi Kenechukwu, John Dike Ogbonna, Chinatu Michael Anyanwu, Nkiru C. Ogbonna, Vivien Ikebudu, Onyinye Precious Chukwudi
article en

Abstract

Accurate prediction of drug release plays an important role in the design of controlled-release pharmaceutical formulations. Traditional formulation development relies heavily on experimental trial-and-error approaches, which are often time-consuming and resource-intensive. This study presents a computational framework for predicting cumulative drug release from key formulation variables and evaluates its performance using a simulated dataset generated with the Korsmeyer-Peppas model. The dataset consisted of 180 simulated formulation samples described by six input variables: polymer concentration, drug load, particle size, dissolution medium pH, dissolution time, and viscosity. Two supervised learning models, Random Forest (RF) and an Artificial Neural Network (ANN), were developed and evaluated using an 80:20 train-test split and five-fold cross-validation. The ANN achieved higher predictive performance than the RF model, with a test R2 of 0.9652, RMSE of 6.855%, and MAE of 6.060%, while the RF model achieved a test R2 of 0.5901, RMSE of 23.539%, and MAE of 16.437%. Feature importance analysis identified particle size and polymer concentration as the most influential variables affecting the predicted drug release profiles. The findings show that machine learning can effectively model relationships between formulation variables and simulated drug release behaviour, and provide a reproducible framework for computational evaluation during the early stages of formulation development. The reported results are based on simulated data and should be interpreted as computational validation of the proposed framework. Future studies will focus on evaluating the framework using experimentally measured dissolution datasets.

Frontiers in Drug DeliveryVol. 6
University of Nigeria (NG), Maduka University (NG)
Advanced Drug Delivery Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon