Development and validation of machine learning models for prediction of nanomedicine solubility in supercritical solvent for advanced pharmaceutical manufacturing
Liu W. Zhao R. Su X. Mohamed A. Diana T.
15 July 2022Elsevier B.V.
Journal of Molecular Liquids
2022#358
Salsalate solubility in supercritical carbon dioxide was studied in this research by computational simulation to correlate the solubility to input parameters including temperature and pressure. A dataset was collected from resources and the models were correlated to the data. Both training and validation steps have been performed to implement the computational tasks. Indeed, we are dealing with a dataset with two P and T inputs and one output (drug solubility), and 32 data points. We chose three models based on Gaussian Process Regression (GPR) including simple (raw) GPR, Ada-boosted GPR, and Bagged GPR as two ensemble methods for correlation of the solubility data. All hyper-parameters were tuned for more general models and models evaluated with standard metrics. The GPR, Adaboost + GPR, and Bagging + GPR have scores of 0.9779, 0.9992, and 0.9795 using R-squared, respectively. Also, in terms of RMSE models have error rates of 1.25 × 10−4, 1.20 × 10−4, and 1.29 × 10−4, respectively. Finally, considering the standard criteria and visual analysis, the boosted GPR model is selected as the main model. The optimal values found as (P = 400, T = 338.0, Y = 0.003879).
Drug manufacture , Drug solubility , Modeling , Nanomedicine , Supercritical process
Text of the article Перейти на текст статьи
Guizhou Academy of Tobacco Science, Guiyang, 550000, China
Research Centre, Future University in Egypt, New Cairo, 11845, Egypt
Department of Technology and Catering Organization, South Ural State University, Chelyabinsk, Russian Federation
Zhangir Khan Agrarian Technical University, Uralsk, Kazakhstan
Guizhou Academy of Tobacco Science
Research Centre
Department of Technology and Catering Organization
Zhangir Khan Agrarian Technical University
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026