Simultaneous Motion Estimation and Image Reconstruction in Cardiac Emission Tomography

Gated cardiac emission computed tomography is a useful procedure for 
diagnosing coronary artery disease. The diagnosis is made on the basis of 
a sequence of three dimensional images of myocardial perfusion or metabolism 
at different phases of the cardiac cycle and on important functional 
parameters such as ejection fraction and myocardial wall motion. The short 
imaging time and cardiac motion reduces the reconstructed image quality and 
introduces uncertainty in the estimated parameters if the images are 
reconstructed independently of each other. As a result, there has been some 
effort to develop image reconstruction techniques which compensate for their 
temporal dependence. On the other hand, the problem of determining the motion 
of the myocardium has been treated using alternate imaging and reconstruction 
methods.

In this talk we discuss how to combine the image reconstruction and wall 
motion estimation in a single algorithm by modeling the heart wall as a 
deformable elastic material.  A novel feature of this method is that it forces
the reconstructed images  to be influenced by the motion estimates. The method
may be viewed as a penalized maximum likelihood image reconstruction method 
with penalty terms determined by (1) an image matching term that ensures 
a measure of agreement between the  gated images and (2)  the strain energy 
of the elastic material model for the heart wall. Simulations will be 
presented to demonstrate the feasibility of the proposed method.