Computer-Assisted Diagnosis and Therapy Planning in Coronary Artery Disease Based on Cardiac CT and MRI

Abstract

State-of-the-art imaging with CT and MRI allows the acquisition of image data for the evaluation of coronary pathologies, myocardial perfusion, detection of infarctions and myocardial function. Thus, information about the cause, effect and potential progression of coronary artery disease is available. The interpretation of the available image data is a difficult task due to the differences in coverage, resolution and contrast types. The goal of the presented work was the development and evaluation of a set of image processing methods to support this task. These methods include registration methods for the alignment of different datasets as well as compensation methods for the motion, which is caused by breathing or myocardial contraction. Other image processing tasks are the segmentation of the coronary arteries and the classification of the myocardium according to its viability status. The proposed methods are integrated into clinically applicable software prototypes and evaluated with phantom and animal datasets as well as with data from clinical routine. The results show a clear benefit for the diagnosis and therapy prognosis of CAD patients based on CT and MRI image data.