Computerized tomographic imaging of time-varying objects using mixture models
R. Deming, LEONID I. PERLOVSKY · 2005
A method is presented for computerized tomographic (CT) imaging of time-varying objects. In this approach, the object func- tion is formulated as a mixture model in which the unknown model parameters describe the local structure, and local motion, of different regions in the object. The Radon transform is used to transform this model into a corresponding mixture model for the set of tomographic projections, and there is a simple relationship between the parameters of the object function and the parameters of the projections. Finally, the problem is solved in an iterative fashion to yield the set of parame- ters giving the best fit between the projection model and the measured projection data. A bonus of the method is the flexibility to process in- complete data in which projections are available only over a limited range of viewing angles. Results of computer simulations are pre- sented to illustrate the algorithm performance. This problem relates to several practical aspects of CT imaging including the correction of motion artifacts caused by patient movement or breathing, as well as the important problem of imaging a beating heart.