Tailor-made MRI for hemodynamic modeling: from artifacts to highly accelerated scans
Personalized models of the cardiovascular system have become increasingly important in recent years. This personalization requires solving an inverse problem based on measured data to adjust parameters used in a mathematical model.
The aim of Miriam Löcke's thesis was to improve the use of magnetic resonance imaging (MRI) data for personalizing these models. This includes analyzing ways to reduce artifacts in the data, integrating MRI modelling with the inverse problem, and finally, using mathematical techniques to find the optimal way to take MRI measurements for this purpose - which can enable an increase in the accuracy of the estimations as well as reducing the amount of scan time, i.e. the amount of time a patient would have to spend inside the machine.