A universal methodology for reliable predicting the non-steroidal anti-inflammatory drug solubility in supercritical carbon dioxide
The article presents a universal methodology for predicting the solubility of non-steroidal anti-inflammatory drugs (NSAIDs) in supercritical carbon dioxide (SCCO2). NSAIDs are a class of drugs that reduce pain, inflammation, and fever. SCCO2 is a solvent that has unique properties such as low viscosity, high diffusivity, and solvating ability.
The article uses different machine learning scenarios to simulate the solubility of twelve NSAIDs in SCCO2 based on their physical characteristics, operating conditions, and solvent property. The article compares the prediction accuracy of twelve intelligent paradigms from three categories: artificial neural networks, support vector regression, and hybrid neuro-fuzzy. The article concludes that adaptive neuro-fuzzy inference is the best tool for the considered task.
The article also compares the proposed hybrid paradigm with the equations of state and available correlations in the literature and finds that it is more reliable. The article provides experimental measurements, model predictions, and relevancy analyses to justify the drug solubility behavior in SCCO2. The article shows that Ibuprofen and Naproxen are the most soluble and insoluble drugs in SCCO2, respectively.
I was involved in the development and implementation of the machine learning scenarios that were used to simulate the solubility of NSAIDs in SCCO2. I also have contributed to the data analysis, model validation, and writing of the article.