Abstract:Over the last few decades, supercritical fluid extraction process has greatly assisted the search for more medicinal and therapeutic properties of bioactive compounds from an extensive range of botanical matrices. The yield of supercritical fluid extraction is often limited by the solubility of a targeted solute in the solvent and thus knowledge on solubility has become one of the fundamental variables of interest in the extraction process. However, the studies to evaluate the performance of density-based empirical models in estimating the solubility of different drug- and plant-based bioactive compounds were scarce. Therefore, this study aimed to gather the solubility data of 33 drugs and 40 plant-based bioactive compounds at varying temperatures and pressures and correlated them with 26 most widely used density-based models. The solubility data were critically analysed and reported from two perspectives: well-studied solubility-influencing factors such as pressure, temperature, and additional influencing factors what o such as molecular structure and weight. Then, a thorough evaluation of the existing empirical solubility models involving 26 predictive models for 73 bioactive compounds was performed. From the results and findings, it was evident that the 8-parameter Belghait et al. model showed the overall highest accuracies indicated by the lowest absolute average relative deviation of 7.42%, 10.16%, and 6.57%, respectively for all bioactive compounds, drug-based bioactive compounds, and plant-based bioactive compounds. Belghait et al. model is also superior since its predictive performances are the best for 54% of the bioactive compounds studied. Meanwhile, Kumar and Johnston's, Sung and Shim's, and Sodeifian et al's models demonstrated the best solubility predictions for 3-, 4-, and 6-parameter empirical models, respectively, for both plant and drug-based compounds. As the effects of pressure and temperature on solubility have widely acknowledged, this study also highlights the influence of molecular structure complexity on solubility. The main findings from this comparative evaluation proved that density-based models are efficacious in predicting solubility of both plant and drug-based bioactive compounds.

A comprehensive study of pharmaceutics solubility in supercritical solvent through diverse thermodynamic and hybrid Machine learning approaches

Recent advancements toward the incremsent of drug solubility using environmentally-friendly supercritical CO2: a machine learning perspective

Optimization of drug solubility inside the supercritical CO2 system via numerical simulation based on artificial intelligence approach

A comprehensive comparison among four different approaches for predicting the solubility of pharmaceutical solid compounds in supercritical carbon dioxide

Machine learning model for prediction of drug solubility in supercritical solvent: Modeling and experimental validation

Application of CO2 Supercritical Fluid to Optimize the Solubility of Oxaprozin: Development of Novel Machine Learning Predictive Models

Prediction of Drug-like Compounds Solubility in Supercritical Carbon Dioxide: A Comparative Study between Classical Density Functional Theory and Machine Learning Approaches

Mathematical modeling and numerical simulation of supercritical processing of drug nanoparticles optimization for green processing: AI analysis

Intelligence computational analysis of letrozole solubility in supercritical solvent via machine learning models

Nonsteroidal anti-inflammatory drug solubility optimization through green chemistry solvent: Artificial intelligence technique

Thermodynamic modeling and solubility assessment of oxycodone hydrochloride in supercritical CO 2 : Semi-empirical, EoSs models and machine learning algorithms

Determination of nanoparticle solubility through green nanonization process using machine learning approach: Computational modeling and optimization

Development a novel robust method to enhance the solubility of Oxaprozin as nonsteroidal anti-inflammatory drug based on machine-learning

Determination of morphine sulfate anti-pain drug solubility in supercritical CO2 with machine learning method

Measurement and modeling of metoclopramide hydrochloride (anti-emetic drug) solubility in supercritical carbon dioxide

Artificial Intelligence aided pharmaceutical engineering: Development of hybrid machine learning models for prediction of nanomedicine solubility in supercritical solvent

Predicting drug solubility in organic solvents mixtures: A machine-learning approach supported by high-throughput experimentation

Empirical Correlations of Drug-and Plant-based Bioactive Compound Solubility in Supercritical CO 2 : A Comparative Evaluation Study

Computational intelligence analysis on drug solubility using thermodynamics and interaction mechanism via models comparison and validation

Medium Gaussian SVM, Wide Neural Network and stepwise linear method in estimation of Lornoxicam pharmaceutical solubility in supercritical solvent

Modeling the solubility of pharmaceuticals in pure solvents and solvent mixtures for drug process design