Electrocardiogram (ECG) is a convenient, economic, and non-invasive detecting tool in myocardial ischemia (MI). And its clinical appearance is mainly exhibited by ST-segment deviation. In the paper, the concepts of Correlation Coefficient Entropy (CCE) and Inverse Correlation Coefficient Entropy (ICCE) were proposed and used to compare the differences in morphology variability between ST segments induced by Heart Rate (HR) and by MI. After the Long-Term ST database (LTST) verification, the obvious results obtained with both methods. Whats more, It showed that CCE was better than ICCE comparatively.
-Level Set Method based on Chan-Vese(C-V) model is widely used in image processing and computer vision. However, there are some drawbacks when C-V model processes Ultrasonic Cardiogram(UCG) images. For example, the accuracy is influenced by noise and speckle in UCG image and some problems such as numerical error and time consuming are caused by re-initialization in level set evolution. Therefore, a novel level set method based on C-V model was proposed in this paper. First of all, the C-V's Partial Differential Equation(PDE) was improved. Second, three signed distance penalizing energy function was analyzed and compared, then the best one which forces the level set function(LSF) to be close to a signed distance function was chosen in this paper. Experiments results showed that the proposed method not only eliminated the effect of speckles on UCG image segmentation, but also reduced the computational cost and avoided numerical errors caused by reinitialization. Besides, the obtained contour curve was much smoother.Index Terms -Ultrasonic Cardiogram (UCG)image segmentation, level set method, Chan-Vese (C-V) model, re-initialization I . IntroductionAs an indispensable inspecting technique for cardiac disease, Ultrasonic Cardiogram(UCG) has several main advantages such as non-invasive, low-price and Real-time. However, because of the high noise, speckle, wide fuzzy boundary region and man-made boundary in UCG, the quality of UCG image is very poor. Especially speckle noise, which makes gray gradient(boundary) occur everywhere in the internal UCG image. Hence, it is more difficult to extract the boundary of UCG than X-CT and MRI image.In the past two decades, active contour models (ACMs)[1] have been widely used in image processing and computer vision, especially for image segmentation. The original ACM was proposed by Kass et al [1]. An energy function, which can restrict close curve, was defined [1]. Then the curve approached the desired boundary by minimizing the energy function. Problems associated with initialization and poor ability to handling topological changes, however, had limited its utility. To eliminate the above drawbacks, Geometric ACM, also called level set method, which implicitly represents the curve by zero level of a high dimensional function. was proposed later by Osher and Sethian [2].Only the local boundary information can be used by the traditional level set method in image segmentation, and it is difficult to acquire desired result when there are fuzzy or discrete boundary region. Therefore, Mumford and Shah[3] proposed the level set method based on Mumford-Shah(M-S) model which used the global information in the homogeneous region. Later, Chan-Vese(C-V) model [4] was proposed by simplifying the M-S model. The C-V model has a perfect utility when disposing fuzzy and discrete boundary which occur in UCG image. However, the C-V model also stops the curve on the speckle boundary but it is not expected. Besides, although re-initialization has been extensively implemented to keep the e...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.