Left ventricle segmentation from cardiac cine MRI to detect cardiac abnormalities
03 Apr 2014-pp 331-334
TL;DR: A simple and novel algorithm is presented to automatically segment the Left Ventricle in short axis cardiac cine MRI to be used as a tool for detecting abnormal cardiac motion.
Abstract: Magnetic Resonance Imaging (MRI) has become one of the most promising technologies for the treatment of cardiac abnormalities. Nowadays, Image processing techniques are widely used in medical field to detect cardiac disorders. In this paper we present a simple and novel algorithm to automatically segment the Left Ventricle (LV) in short axis cardiac cine MRI. The frames of the cine MRI of a pathological and five normal patients are processed using enhancement, thresholding and morphological operations. The variation in the area of the left ventricle is calculated for the entire cardiac cycle for normal and diseased conditions. The pattern of variation in area for abnormal is significantly different from that of normal and hence could be used as a tool for detecting abnormal cardiac motion.
Citations
More filters
01 Jan 2020
TL;DR: Estimation of the movement of the LV shows that the cardiac motion varies significantly in normal and abnormal LV.
Abstract: Echocardiographic images are widely used in the diagnostic procedure of the cardiac function. Left ventricle plays a vital role in pumping the oxygenated blood to the complete body and maintain systematic circulation. In this study, the echocardiogram data of a normal and abnormal subject were collected. The frames were extracted from the video for one cardiac cycle. The left ventricle from each frame was segmented using image processing techniques. The parameters such as area, perimeter, and the centroid of the left ventricle were determined. These parameters were used to estimate the movement of the LV and measure the contraction and expansion of the chamber while pumping the blood out in one cardiac cycle. The comparison of these parameters in normal and abnormal LV show that the cardiac motion varies significantly. ECG signal was used as the biomarker for this estimation.
References
More filters
TL;DR: This paper proposes an original categorization for cardiac segmentation methods, with a special emphasis on what level of external information is required (weak or strong) and how it is used to constrain segmentation.
703 citations
"Left ventricle segmentation from ca..." refers background in this paper
...Today, cardiac MRI is recognized as the modality for the assessment of left ventricular function [2]....
[...]
...Usually in order to calculate the area and other features of the heart, doctors have to draw contours manually, slice by slice which is very tedious and often delay the whole diagnosis procedures [2]....
[...]
TL;DR: The developments directed toward automated quantitative image analysis and semi‐automated contour detection for cardiovascular MR imaging are reviewed.
Abstract: Magnetic resonance imaging (MRI) offers several acquisition techniques for precise and highly reproducible assessment of global and regional ventricular function, flow, and perfusion at rest and under pharmacological or physical stress conditions. Recent advances in hardware and software have resulted in strong improvement of image quality and in a significant decrease in the required imaging time for each of these acquisitions. Several aspects of heart disease can be studied by combining multiple MRI techniques in a single examination. Such a comprehensive examination could replace a number of other imaging procedures, such as diagnostic X-ray angiography, echocardiography, and scintigraphy, which would be beneficial for the patient and cost effective. Despite the advances in MRI, quantitative image analysis often still relies on manual tracing of contours in the images, which is a time-consuming and tedious procedure that limits the clinical applicability of cardiovascular MRI. Reliable automated or semi-automated image analysis software would be very helpful to overcome the limitations associated with manual image processing. In this paper the developments directed toward automated quantitative image analysis and semi-automated contour detection for cardiovascular MR imaging are reviewed. J. Magn. Reson. Imaging 1999; 10:602–608. © 1999 Wiley-Liss, Inc.
192 citations
TL;DR: An updated and critical review of cardiac motion tracking methods including major references and those proposed in the past ten years is provided and can serve as a tutorial for new researchers entering the field.
Abstract: Magnetic resonance imaging (MRI) is a highly advanced and sophisticated imaging modality for cardiac motion tracking and analysis, capable of providing 3D analysis of global and regional cardiac function with great accuracy and reproducibility. In the past few years, numerous efforts have been devoted to cardiac motion recovery and deformation analysis from MR image sequences. Many approaches have been proposed for tracking cardiac motion and for computing deformation parameters and mechanical properties of the heart from a variety of cardiac MR imaging techniques. In this paper, an updated and critical review of cardiac motion tracking methods including major references and those proposed in the past ten years is provided. The MR imaging and analysis techniques surveyed are based on cine MRI, tagged MRI, phase contrast MRI, DENSE, and SENC. This paper can serve as a tutorial for new researchers entering the field.
146 citations
01 Feb 1971
107 citations
"Left ventricle segmentation from ca..." refers methods in this paper
...The conventional edge detection techniques like Roberts, Sobel, Prewitt, Laplacian, and Canny, based on the difference of gray levels, failed to extract the correct boundaries of the MR images [5]....
[...]
11 Nov 2008
TL;DR: In this article, the authors presented a novel edge following technique for image segmentation designed to segment the left ventricle in cardiac magnetic resonance (MR) images, which is a required step to determine the volume of left ventricular volume in a cardiac MR image.
Abstract: This paper presents a novel edge following technique for image segmentation designed to segment the left ventricle in cardiac magnetic resonance (MR) images. This is a required step to determine the volume of left ventricle in a cardiac MR image, which is an essential tool for cardiac diagnosis. The traditional method for extracting them from cardiac MR images is by human delineation. This method is accuracy but time consuming. So a new ventricular segmentation technique is proposed in order to reduce the analysis time with similar accuracy level compared to doctorspsila opinions. Our proposed technique can detect ventricle edges in MR images using the information from the vector image model and the edge map. We also compare the proposed segmentation technique with the active contour model (ACM) and the gradient vector flow (GVF) by using the opinions of two skilled doctors as the ground truth. The experimental results show that our technique is able to provide more accurate segmentation results than the classical contour models and visually close to the manual segmentation by the experts. The results evaluated using a numerical measure by mean of the probability of error in image segmentation also confirm the visual evaluation.
4 citations