The analysis of cardiac function in patients suffering from arrhythmia poses a problem for conventional ECG-synchronized imaging and the patients' ability to hold their breath. Real-time imaging approaches provide ungated image data, which contains the information about the motion variation induced by breathing and arrhythmia but require a high effort in post-processing and interpretation. The goal of the presented work is to enable an automatic analysis of cardiac real-time image sequences of patients suffering from arrhythmia. To this end, we combine a fast CNN-based segmentation of the myocardium with a curve pattern analysis of the blood volume changes over time.
This abstract and the presentation materials are available to members only; a login is required.