Cardiac magnetic resonance imaging (MRI) has become an indispensable diagnostic tool in cardiovascular medicine due to its high spatial resolution and available tissue contrasts. Despite all the advantages, the procedures used in clinical practice are often slow, not very robust, and not comfortable for patients (repeated breath holding). Because of motion artefacts and arrhythmic heartbeats, they therefore do not always provide optimal diagnostic information. In recent years, we have developed a method for real-time MRI that eliminates these disadvantages (Uecker et
al., 2010b). However, this method still has some limitations, for example, it is limited to imaging two-dimensional slices. In many cases, however, the precise clinical assessment of a clinical picture of the heart requires 4D information, i.e. time-resolved imaging of a three-dimensional volume that encompasses the entire heart.
heart. Although it is also possible to achieve coverage of the heart by measuring a stack of 2D slices, these images can then no longer be easily assigned in terms of time in a real-time measurement and, in contrast to a 3D volume, do not allow subsequent reformatting into arbitrary planes. A higher spatial resolution is also desirable than is currently possible with real-time MRI.