3D Cardiac Modeling Based Medical Images And Functional Parameters Analysis

Three-dimensional (3-D) imaging of the heart is has developedrapidly in research area of medical imaging. Many scientists havepresented various approaches for the analysis of the cardiac shape andparameters based on the development of new classes of left ventricularmodels. Currently, the deformable models have become an important toolfor the medical image analysis. It incorporates the data constraints derivedfrom images and prior knowledge into the variational framework and itsapplication has covered image denoising, image segmentation, imageregistration and motion tracking. Particularly, techniques based onvolumetric deformable models and spatio-temporal geometric models havereceived considerable attention. From this paper, we firstly apply a newtechnique based on a 3D-ASM, which is presented for automaticsegmentation of cardiac SPECT image data sets in our country. In thispaper, we can conclude that whereas 3-D statistical model-basedapproaches have the capability to improve the diagnostic value of cardiacimages, such as robustness, 3-D interaction, computational complexity andclinical validation still require significant attention. Furthermore, we visualise that parameters based on the statistical models is useful tophysicians. Thus a new approach is presented for analyzing the motion ofleft ventricle (LV) from SPECT (Single Photon Emission ComputedTomography) imaging data. The technique applied in this paper is based ona statistical model which is one of special of physics-based deformablemodels. By means of 3D Active Shape Model (3D-ASM) algorithm, endoand epicardial borders of the LV can be obtained, thus a geometrical andgrey-level statistical models based a training set of cardiac SPECT studiescan be built. Finally, the deformation and complex motion of the LV areestimated in terms of a few of parameters that vary during a cycle (700ms).With the help of these parameters, we can compute the volume and EF ofcardic, and physicians can characterize the motion of the LV in a clinicallyway.