Advanced modeling based on machine learning for evaluation of drug nanoparticle preparation via green technology: Theoretical assessment of solubility variations

Computational intelligence modeling of hyoscine drug solubility and solvent density in supercritical processing: gradient boosting, extra trees, and random forest models

“…The given dataset comprises of 45 instances that represent the solubility of the Hyoscine drug at distinct combinations of temperature and pressure. The input variables considered for the dataset are temperature in Kelvin and pressure in bar, whereas the output variables are density and solubility 3 . The entire data set is displayed in Table 1 which has been obtained from 37 .…”

“…Estimating pharmaceutical solubility in supercritical solvents such as CO 2 has been reported by different methods such as thermodynamics and data-driven models 3 . The main inputs for the modeling have been considered to be pressure and temperature as these factors showed the most important effects on the drug solubility change 2 , 4 – 7 .…”

“…The methods of machine learning can be employed for both solubility prediction and solvent density, as the solvent is considered to be compressible in this process. Begum used several methods including SVM, KNN, and linear regression (LR) to predict the solubility of Hyoscine and density of solvent in supercritical processing, and the results were promising 3 . However, other methods based on machine learning can be developed and tested for correlation of Hyoscine solubility in supercritical CO 2 solvent.…”

“…The dataset used in this study contains 45 observations of Hyoscine solubility and density of the solvent at various combinations of temperature and pressure. The input variables are temperature (in K) and pressure (in bar), while the output variables are solubility (mole fraction) and density (kg m -3 ) 3 . The dataset was preprocessed using min–max scaling to normalize the input variables.…”

Finding the Right Solvent: A Novel Screening Protocol for Identifying Environmentally Friendly and Cost-Effective Options for Benzenesulfonamide

“…Extensive experimental screening, although the most reliable, is limited due to the time, effort, and costs needed. Hence, machine learning offers a real alternative for exploring the solvent space, provided that a reliable model has been developed [ 39 , 40 , 41 , 42 ]. As was established in previous studies [ 42 , 43 ], combining quantum chemical methods, such as COSMO-RS (Conductor-like Screening Model for Real Solvents) with machine learning methods, is a quite promising approach, providing good-quality predictions.…”