Studies from National and Kapodistrian University of Athens Reveal New Findings on Machine Learning (Simulating Realistic Patient Profiles From Pharmacokinetic Models By a Machine Learning Postprocessing Correction of Residual Variability).

A report from the National and Kapodistrian University of Athens discusses research on machine learning in pharmacokinetics. The researchers developed a method to address model misspecification in population pharmacokinetics by using machine learning to model residual unexplained variability. The methodology was tested in a study of ropinirole and was found to generate realistic virtual patient profiles and quantify the extent of model misspecification. The research concludes that the greater the extent of variability explained by the machine learning model, the higher the degree of model misspecification in the original model. [Extracted from the article]