Motion-guided segmentation for cine DENSE MRI

Abstract

Defining myocardial contours is often the most time-consuming portion of dynamic cardiac MRI image analysis. Displacement encoding with stimulated echoes (DENSE) is a quantitative MRI technique that encodes tissue displacement into the phase of the complex MRI images. Cine DENSE provides a time series of these images, thus facilitating the non-invasive study of myocardial kinematics. Epicardial and endocardial contours need to be defined at each frame on cine DENSE images for the quantification of regional displacement and strain as a function of time. This work presents a reliable and effective two-dimensional semi-automated segmentation technique that uses the encoded motion to project a manually-defined region of interest through time. Contours can then easily be extracted for each cardiac phase. This method boasts several advantages, including, (1) parameters are based on practical physiological limits, (2) contours are calculated for the first few cardiac phases, where it is difficult to visually distinguish blood from myocardium, and (3) the method is independent of the shape of the tissue delineated and can be applied to short- or long-axis views, and on arbitrary regions of interest. Motion-guided contours were compared to manual contours for six conventional and six slice-followed mid-ventricular short-axis cine DENSE datasets. Using an area measure of segmentation error, the accuracy of the segmentation algorithm was shown to be similar to inter-observer variability. In addition, a radial segmentation error metric was introduced for short-axis data. The average radial epicardial segmentation error was 0.36 ± 0.08 and 0.40 ± 0.10 pixels for slice-followed and conventional cine DENSE, respectively, and the average radial endocardial segmentation error was 0.46 ± 0.12 and 0.46 ± 0.16 pixels for slice following and conventional cine DENSE, respectively. Motion-guided segmentation employs the displacement-encoded phase shifts intrinsic to DENSE MRI to accurately propagate a single set of pre-defined contours throughout the remaining cardiac phases.

Introduction

A number of MRI techniques have been developed to quantify myocardial motion, including myocardial tagging (Zerhouni et al., 1988, Axel and Dougherty, 1989), phase contrast (PC) velocity encoding (van Dijk, 1984, Bryant et al., 1984), and more recently, displacement encoding with stimulated echoes (DENSE) (Aletras et al., 1999). DENSE has the advantage over velocity-encoded PC of directly measuring tissue displacement instead of velocity, and the advantages over tagging of higher spatial resolution and more direct computation of displacement. Quantitative methods are potentially useful for reducing subjectivity and improving accuracy (Gotte et al., 2001) in the clinical assessment of cardiac wall motion, but their clinical use is currently limited by lack of automation.

Defining epicardial and endocardial contours is an integral step in quantifying regional cardiac wall motion. For cine DENSE (Kim et al., 2004) these contours are typically manually delineated for all cardiac phases, which is a laborious process and is currently the most time-consuming component of the cine DENSE image analysis. Automated myocardial contour detection techniques based on image intensity may not be well-suited to cine DENSE because: (1) boundaries between the myocardium and adjacent tissue (e.g. the liver) are often indiscernible based on signal magnitude; (2) a T1-related decay in signal-to-noise ratio (SNR) with time is often present; and (3) high signal is present in the blood pool of the first few frames, before it is washed out of the image plane.

A number of advances have been made towards automating image segmentation for myocardial tagging and velocity encoding (Kumar and Goldgof, 1994, Young et al., 1995, Kraitchman et al., 1995, Guttman et al., 1994, Montillo et al., 2002, Montillo et al., 2003, Wong et al., 2002, Cho and Benkeser, 2006) but no segmentation algorithm specifically tailored for cine DENSE has been developed. A segmentation method is presented here that uses tissue tracking based on the motion encoded into the phase of the cine DENSE images themselves to project individual manually-defined contours through time. The method is shown to be accurate, versatile and practical, and forms a significant step towards the automation of cine DENSE image analysis.