Related Experiment Video
Updated: Mar 16, 2026
09:51
Microfluidic Production of Lysolipid-Containing Temperature-Sensitive Liposomes
Published on: March 3, 2020
10.0K
Prediction and Classification of Liposomal Release and Stability Using Machine Learning Based on Ethanol and Tergitol
Jeong Seon Hwang1, Panalee Pomseethong1, Jin-Chul Kim1
1Department of Biomedical Science & Institute of Bioscience and Biotechnology, Kangwon National University, Chuncheon, 24341, Republic of Korea.
Current Pharmaceutical Design
|March 15, 2026
Summary
Tergitol surfactant concentration significantly impacts liposome release more than ethanol. Machine learning models accurately predict liposome stability and release behavior for pharmaceutical and cosmetic formulations.
Area of Science:
- Pharmaceutical Sciences
- Materials Science
- Computational Chemistry
Background:
- Liposomes are versatile drug delivery vehicles with excellent biocompatibility.
- Formulation additives like ethanol and surfactants can compromise liposome stability and alter drug release.
- Understanding these interactions is crucial for developing stable and effective liposomal products.
Purpose of the Study:
- To develop predictive models for liposomal release and stability.
- To quantify the impact of ethanol and Tergitol surfactants on liposome behavior.
- To apply machine learning for optimizing liposome formulation design.
Main Methods:
- Multiple linear regression models were used to predict drug release based on ethanol and Tergitol concentrations.
- Nonlinear interactions were visualized using 3D regression surfaces.
- K-nearest neighbors, logistic regression, and stochastic gradient descent classified liposome stability.
Main Results:
- Regression models showed high accuracy (R² 0.9611-0.9899) in predicting liposomal release.
- Logistic regression yielded the highest accuracy (87.98%) in classifying liposome stability.
- Tergitol concentration demonstrated a more significant effect on release than ethanol.
Conclusions:
- Tergitol concentration is a key factor influencing liposome release, with higher HLB surfactants showing reduced interaction.
- The developed machine learning models offer practical tools for predicting formulation outcomes.
- These models can aid in the rational design of liposome-based pharmaceutical and cosmetic products.
