Myocardial border detection by branch-and-bound dynamic programming in magnetic resonance images

Yeh, Jinn-Yi; Fu, Jachih; Wu, Chia-Chang; Lin, Hu; Chai, Jyh‐Wen

doi:10.1016/j.cmpb.2004.10.010

Cited by 30 publications

(17 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image-based methods propose to process them differently and separately, hence the focusing of certain methods on the LV endocardium only. The first step consists in finding the endocardial contour, usually with thresholding (Goshtasby and Turner, 1995;Weng et al, 1997;Nachtomy et al, 1998;Katouzian et al, 2006) and/or dynamic programming (DP) (Gupta et al, 1993;van der Geest et al, 1994;Geiger et al, 1995;Lalande et al, 1999;Fu et al, 2000;Yeh et al, 2005;Liu et al, 2005;Uzümcü et al, 2006). DP applied to image segmentation consists in searching for the optimal path in a cost matrix that assigns a low cost to object frontiers.…”

Section: Image-based Methodsmentioning

confidence: 99%

“…Here, the circular geometry of the LV is taken advantage of using polar coordinates, so as to make the search problem one-dimensional. The design of the cost matrix, which is the main difficulty in this problem, can be based on thresholding (van der Geest et al, 1994;Liu et al, 2005), fuzzy logic (Lalande et al, 1999), gray levels, using wavelet-based enhancement (Fu et al, 2000) or radial lines (Yeh et al, 2005), or on the gradient values, used in (Cousty et al, 2010) to weight a spatio-temporal graph. In Jolly et al (2009), a shortest path algorithm is applied on an image obtained by averaging all the phases over one cardiac cycle, and contours in each individual image are recovered using minimum surface segmentation.…”

Section: Image-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A review of segmentation methods in short axis cardiac MR images

Petitjean

Dacher

2011

Medical Image Analysis

633

431

View full text Add to dashboard Cite

For the last 15 years, Magnetic Resonance Imaging (MRI) has become a reference examination for cardiac morphology, function and perfusion in humans. Yet, due to the characteristics of cardiac MR images and to the great variability of the images among patients, the problem of heart cavities segmentation in MRI is still open. This paper is a review of fully and semi-automated methods performing segmentation in short axis images using a cardiac cine MRI sequence. Medical background and specific segmentation difficulties associated to these images are presented. For this particularly complex segmentation task, prior knowledge is required. We thus propose 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. After reviewing method principles and analyzing segmentation results, we conclude with a discussion and future trends in this field regarding methodological and medical issues.

show abstract

Section: Image-based Methodsmentioning

confidence: 99%

Section: Image-based Methodsmentioning

confidence: 99%

A review of segmentation methods in short axis cardiac MR images

Petitjean

Dacher

2011

Medical Image Analysis

633

431

View full text Add to dashboard Cite

show abstract

“…3, in the last layer, a fully connected layer containing one node was added for final output. We resized the input image into 224*224 and adopted the uniform weight to initialize the first convolutional layer and the last FC layer (because the number of input channels is different from the traditional RGB images, and the last FC layer does not exist in the natural images classification tasks), and the other layers were initialized by the pre-trained CNN model on ImageNet datasets [40] The loss function for optimization is the formula (8) (8) where N refers to the number of training examples, X i denotes prediction value of the ith example, and Y i denotes the ground truth of the ith example. In this section, we trained the networks using Adam optimization method [41].…”

Section: Cnn Designmentioning

confidence: 99%

“…However, automatic LV segmentation is still an open and challenging task, due to the inherent characteristics of the cardiac MR images [4], such as higher noise, intensity level inhomogeneity, effect of partial volume [5,6], complex topological structures, and great variability across different slices. Before 2011, the LV segmentation methods can be categorized into four kinds: 1) The methods based on traditional image-driven technologies, e.g., threshold based methods [5,7], dynamic programming based methods [8], registration based methods [9], and graph based methods [10].…”

mentioning

confidence: 99%

Multi-Views Fusion CNN for Left Ventricular Volumes Estimation on Cardiac MR Images

Luo¹,

Dong²,

Wang

et al. 2018

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

show abstract

“…Histogram equalization (HE) [3,5,7], adaptive HE (AHE) [11,16,18], and weighted histogram separation (WHS) [15] are widely used statistical domain methods. Spatial domain methods operate directly on pixels, such as border detection [1,22] and noise reduction [8]. Frequency domain methods, which apply the Fourier transform of an image, are used to enhance the texture with a repeated pattern [2].…”

Section: Introductionmentioning

confidence: 99%

Image enhancement of shoulder magnetic resonance imaging in rotator cuff injury diagnosis

Lin

et al. 2012

Journal of the Chinese Institute of Industrial Engineers

View full text Add to dashboard Cite

Purpose: Magnetic Resonance Imaging is the most common tool used to diagnose supraspinatus muscle rupture or tear. However, discriminating the features between the rupture and the tear in MR images is difficult, and leads to low diagnostic accuracy. In this article, image enhancement techniques are performed on the original images to improve diagnostic performance. Methods: Three types of histogram-based image enhancement techniques are employed to enhance rotor cuff MR images: (1) histogram equalization; (2) weighted histogram separation (WHS); and (3) coupled WHS. The peak signal-to-noise ratio (PSNR) was used to evaluate the quality of the enhanced images. The experimental group comprises the enhanced images with the highest PSNR and the original MR images. The control group comprises only the original MR images. Both the control group and experimental group were randomly presented to three orthopedic physicians for performance evaluation. The accuracy rate and F-measure were applied to measure the average classification performance. The McNemar test was then employed to test the significance levels of the proposed enhancement techniques. Results: Experimental results showed the original MR image combined with the coupled WHS images (experimental group) outperforms the original MR image (control group) both in accuracy rate and F-measure. The McNemar test shows that the experimental group significantly outperforms the control group in terms of classifying the injury by type. Conclusion: A coupled WHS image enhancement technique was developed to assist the diagnosis of rotator cuff MR imaging by type. Experimental results showed that the proposed enhancement techniques improve the average diagnostic performance and facilitate the classification of the injury by type.

show abstract

Myocardial border detection by branch-and-bound dynamic programming in magnetic resonance images

Cited by 30 publications

References 22 publications

A review of segmentation methods in short axis cardiac MR images

A review of segmentation methods in short axis cardiac MR images

Multi-Views Fusion CNN for Left Ventricular Volumes Estimation on Cardiac MR Images

Image enhancement of shoulder magnetic resonance imaging in rotator cuff injury diagnosis

Contact Info

Product

Resources

About