Model-based trajectory optimization of treatment regimens in chronic diseases
Chronic diseases require long-term treatments that are continuously adapted to current physiological requirements. Often, a compromise must be found between desired treatment outcomes and side effects. For example, chemotherapies have to fulfill the subtle compromise between inducing cell death in tumor cells and potentially life-threatening hematotoxicity. To improve protocols for optimal interference with biomedical systems, my team and I have been working on methods for model-guided trajectory optimization using machine-learning techniques such as reinforcement learning. We hypothesize that current biochemical knowledge about the pathways affected by pharmacological treatments can be used for optimizing the administration dynamics to increase their therapeutic efficiency. I will start with application examples in cellular systems that could be experimentally validated. Subsequently, I will present application examples in chronic diseases.
Announcement of the SysBio Talk of Dr. Dr. Stefan Kallenberger