Conference Paper

Semi automatic estimation and visualization of left ventricle volumes in cardiac MRI

ETSI Telecomunicacion, Univ. Politecnica de Madrid
DOI: 10.1109/CIC.2005.1588121 Conference: Computers in Cardiology, 2005
Source: IEEE Xplore


The main goal of this work was to implement an easy and convenient new tool to load, visualize and segment cardiac magnetic resonance images (CMRI) in order to obtain basic parameters for cardiac function estimation and to visualize the results in a quantitative and graphical way. The proposed method tracks the contour of the left ventricle at endocardial level through all the cardiac cycle to measure the ventricular volume and the ejection fraction, using active contour techniques to differentiate structures within an image

Download full-text


Available from: Norberto Malpica,
  • Source
    • "Moreover the important variation of size and shape of a pathological myocardium represents a challenge of the model construction. Deformable models have also been widely used for segmenting cardiac images [7], [8], [9], [10], [11]. Active contour models have been quite successful for segmenting the myocardium boundaries using CMR images. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we present an automatic approach to segment Cardiac Magnetic Resonance (CMR) images. A preprocessing step that consists in filtering the image using connected operators (area opening and closing filters) is applied in order to homogenize the cavity and solve the problems due to the papillary muscles. Thereby the GVF snake algorithm is applied with one point clicked in the cavity as initialization and an optimized tuning of parameters for the endocardial contour extraction. The epicardial border is then obtained using the endocardium as initialization. The performance of the proposed method was assessed by experimentation on thirty-nine CMR images. A high agreement between manual and automatic contours was obtained with correlation scores of 0.96 for the endocardium and 0.90 for the epicardium. Overlapping percentage, mean and maximum distances between the two contours show a good performance of the method.
    Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 02/2007; 2007:4508-11. DOI:10.1109/IEMBS.2007.4353341
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cardiac magnetic resonance imaging (MRI) is one of the essential non-invasive modalities for the diagnosis of cardiovascular diseases. The main advantage of MRI compared to other techniques is its ability to study in only one examination the contraction and the viability. Left ventricle segmentation is very important to follow-up ischemic diseases. A novel approach to detect endocardial border on cardiac magnetic resonance (CMR) images is presented in this paper. It consists in filtering short axis CMR images, using connected operators (area-open and area-close filters) to homogenize the cavity, prior to the segmentation which is performed using GVF-snake algorithm in two steps. This method does not require a complicated parameters adjustment. Validation was performed on thirty-nine slices by comparing resulting segmentation to the manual contours traced by an expert. This comparison showed good results with an overall average similarity area of 90.7% and an average distance between the two contours of 0.6 pixel. Thus this method can be proposed for clinical evaluation.
    IRBM 09/2007; 28(3):117-123. DOI:10.1016/j.rbmret.2007.07.001 · 0.52 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Based on discrete mathematical morphology, we introduce in this paper a new watershed framework which allows for segmenting spatio-temporal images, that we apply to medical image analysis. Specifically, we propose a new automated and fast procedure to segment the left ventricular myocardium in 4D (3D+t) cine-MRI sequences. Both quantitative and qualitative evaluation are provided. Thanks to the comparison with manual segmentation performed by two cardiologists, we demonstrate the accuracy of the proposed method. The relevance of the ejection fraction and myocardium mass measured from segmentations is also assessed. Furthermore, we show that the proposed 4D procedure allows to keep the temporal coherency between the successive 3D segmentations obtained along the time sequence. In an effort to promote open science, both the data and the software are available on-line.
    Image and Vision Computing 08/2010; 28(8-28):1229-1243. DOI:10.1016/j.imavis.2010.01.001 · 1.59 Impact Factor
Show more