Segmentation of Cardiac Structures Simultaneously from Short- and Long-Axis MR Images

We introduce a framework for the automatic segmentation of the ventricles, atria, and epicardium simultaneously from cardiac magnetic resonance (MR) volumes. The basic idea is to utilize both short-axis (SA) and long-axis (LA) MR volumes. Consequently, anatomical information is available from the whole heart volume. In this paper, the framework is used with deformable model based registration and segmentation methods to segment the cardiac structures. A database consisting of the cardiac MR volumes of 25 healthy subjects is used to validate the methods. The results presented in this paper prove that by using both the SA and LA MR volumes the ventricles and atria can be simultaneously segmented from cardiac MR volumes with a good accuracy. The results show that notably better segmentation results are obtained when the LA volumes are used in addition to the SA volumes. For example, this enables accurate segmentation of the ventricles also in the basal and apical levels.