Data Analysis and Statistical Modelling of Dynamical Processes in Neurology GEM mit TI 315/4-1

点击翻译按钮获取中文摘要

Numerous clinical phenomena in neurology are of dynamic oscillatory nature and are caused by interacting sub-systems. The analysis of multivariate time series of the dynamics of the subsystem allows for an understanding of the underlying processes in the healthy and the diseased state. The aim of this project is to develop and apply mathematical methods in order to infer from time series the dynamical properties of the subsystem and their interactions. There are different modes of interaction, ranging from signal propagation via resonance to synchronization. Methods will be developed to discriminate between these different types of relationship for bi- and multivariate time series. In close collaborations with clinicians, the methods will be tested for their appropriateness and utilized in basic and clinical research in three neurologic fields, i.e. generating mechanisms of pathological tremors, prediction of epileptic seizures, and diagnosis of cerebrovascular diseases. The development of the mathematical methods will be driven by the challenges in data analysis posed by the clinical applications. On the other land, the application of the developed methods are expected to enhance clinical understanding, provide diagnostic and prognostic factors and eventually guide treatment strategies.