Dynamic Heart Modeling Based on a Hybrid 3D Segmentation Approach

A hybrid 3D segmentation approach is proposed in this paper to perform a physical beating heart modeling from dynamic CT images. A Morphological Recursive Erosion operation is firstly employed to reduce the connectivity between the heart and its neighborhood; then an improved Fast Marching method is introduced to greatly accelerate the initial propagation of a surface front from the user defined seed structure to a surface close to the desired heart boundary; a Morphological Reconstruction method then operates on this surface to achieve an initial segmentation result; and finally Morphological Recursive Dilation is employed to recover any structure lost in the first stage of the algorithm. Every one of 10 heart volumes in a heart beating cycle is segmented individually and finally aligned together to produce a physical beating heart model. This approach is tested on 5 dynamic cardiac groups, totally 50 CT heart images, to demonstrate the robustness of this technique. The algorithm is also validated against expert identified results. These measurements revealed that the algorithm achieved a mean similarity index of 0.956. The execution time for this algorithm extracting the cardiac surface from a dynamic CT image, when run on a 2.0 GHz P4 based PC running Windows XP, was 36 seconds.