Parts | Days | Selection | Search | Updates | Downloads | Help
CPP: Fachverband Chemische Physik und Polymerphysik
CPP 2: Modeling and Simulation of Soft Matter I
CPP 2.1: Invited Talk
Monday, March 27, 2023, 09:30–10:00, MER 02
Molecular Theories meet Explainable Machine Learning – Novel Concepts for Advanced Drug Formulations — •Jens Smiatek — Institute for Computational Physics, University of Stuttgart, Germany — Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach (Riss), Germany
Pharmaceutical processes and drug formulations are based on a variety of molecular mechanisms and principles. Significant advances have been made in recent years in terms of basic understanding. Despite these advances, a large number of mechanisms are not yet fully understood, which would be of great advantage, especially for the optimization of development processes as well as the quality and stability of drugs. In this talk I will present fundamental molecular theories of electronic interactions in solutions. Due to the large number of components and the complexity of the interactions, these theories can only be used for a qualitative understanding. However, by combining it with explainable machine learning, the basic molecular mechanisms can be identified and the corresponding thermodynamic properties of the solutions can be predicted. I will present some examples and discuss the underlying benefits and additional challenges for future developments.