Atrial Fibrillation (AF) is a growing problem in modern societies with an enormous impact in both short term quality of life and long term survival. A recently developed promising approach to cure AF uses radio frequency (RF) ablation to carry out "pulmonary vein antrum isolation" (PVAI) to the heart. However, the lack of proper 3D visualization during surgery training, planning, and guidance makes surgery a very difficult task to the surgeons and therefore the risk for the patient increases.
The purpose of this work is to develop methods for automatically segmenting and tracking the heart in 4-D cardiac MRI datasets. The heart surface will be reconstructed to serve as a virtual computer model for the 3D training, planning and guidance stages of cardiac surgery.
Our method is based on an active contour model and a geometric post-processing
of the segmentation to enforce temporal and spatial consistency.
In a first stage the heart region is automatically localized by
making use of the time variance of the heart region. Extraction
of heart border is performed by means of energy minimization.
Finally, we construct a 3D tensor array from all time surfaces
in cylindrical coordinates and time and space consistency of the
segmentation is enforced by fitting an m-variate tensor smoothing
spline to the final 3D array.
