Comparative Study of Left Ventricular Low Wall Motion with Scar Tissue Using 4D Left Ventricular Cardiac Images

Abstract

Myocardial contraction affects the cardiovascular pumping system, and helps in the early phase to detect abnormalities of wall motion noninvasively. In this research, we designed a program to characterize regional abnormalities because scar tissue is very difficult to identify in normal cardiac CT images. We created 10 frames of a 3D heart model that contains the long axis as reference for predicting the left ventricular wall motion. We tested our 4D cardiac model with scar tissue using non invasive cardiac CT images. Here, four subjects (patients) were involved in this study. Subject 1 and 4 are matching the low motion of surgical area with scar tissue area. Subject 2 found fibrous tissue regions (about 40%), compared with the 2SD (Standard Deviation) region. The fibrotic area is completely overlapped with a low-motion region which indicates the fibrotic area has a significant correlation with the low wall motion region. This research evaluates low wall motion of the left ventricle and detection of fibrosis regions.Keywords: Left ventricular remodeling; myocardial wall motion; Cardiac imaging; Fibrotic tissue; Noninvasive cardiac evaluation; IntroductionCardiac disease has become a very challenging clinical problem like regional myocardial injury. It is one of the goals of cardiac imaging methods to measure the regional function of the left ventricle (LV). There are many existing techniques to precise and reliable quantitative regional LV function measurements. Most of the standard methods depend on 2D image sequence data [1,2,3,4,5,6].Most of the techniques have been used end-diastolic(ED) and end systolic (ES) image frames. While the LV expansion and LV thickening from region to region is indicative of ischemia [2,4,6]. Song and Leahy have performed the dense-field optical flow approach to include fluid flow models using 3D datasets [7]. The Goldgof research group followed a shape matching idea which is similar to ours; though they primarily use Gaussian curvature employing conformal stretching models [8]. Pentland and Terzopoulos have been studying non-rigid motion models, using finite element analysis, which might be useful for cardiac analysis. The Ayache group unified these two approaches to segment and track the object simultaneously [9]. A recently proposed technique is the use of phase contrast MR images to decipher local velocity [10]. However, MR images take a long in acquisition and resolution is less than CT images. This can be integrated to estimate trajectories of individual points over time [11,12]. Our new