Cardiac MR image segmentation using deformable models

Gupta, Alok; Kurowski, Lorenz von; Singh, Ajit Pratap; Geiger, Davi; Liang, Cheng-Chung; Chiu, Ming-Yee; Adler, Lee P.; Haacke, Mark; Wilson, David L.

doi:10.1109/cic.1993.378377

Cited by 34 publications

(22 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, the evolution took about 67 steps which ran for 8 s. It is very interesting to compare the contour extraction results in Figs. 1-4 with similar MR cardiac images using standard snakes, e.g., in [20]. Using the traditional methodology one must start quite close to the contour of interest in order to be able to capture in.…”

Section: A Contour Extraction Resultsmentioning

confidence: 99%

“…Assume that is negative in the interior and positive in the exterior of the zero level set. We consider the zero level set, defined by (17) We have to find an evolution equation of , such that the evolving curve is given by the evolving zero level , i.e., (18) By differentiating (17) with respect to we obtain (19) Note that for the zero level, the following relation holds: (20) In this equation, the left side uses terms of the surface , while the right side is related to the curve . The combination of (16) to (20) gives (21) and the curve , evolving according to (16), is obtained by the zero level set of the function , which evolves according to (21).…”

Section: B Level Set Representationsmentioning

confidence: 99%

“…For example, snakes have been employed for the segmentation of myocardial heart boundaries as a prerequisite from which such vital information such as ejection-fraction ratio, heart output, and ventricular volume ratio can be computed. (See [20], [42], and the references therein.) In this paper, we will apply a new snake paradigm which the authors have developed [25], [26] for edge detection and segmentation of various kinds of medical imagery including magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A geometric snake model for segmentation of medical imagery

Yezzi

Kichenassamy

Kumar

et al. 1997

IEEE Trans. Med. Imaging

487

299

View full text Add to dashboard Cite

Abstract-In this note, we employ the new geometric active contour models formulated in [25] and [26] for edge detection and segmentation of magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound medical imagery. Our method is based on defining feature-based metrics on a given image which in turn leads to a novel snake paradigm in which the feature of interest may be considered to lie at the bottom of a potential well. Thus, the snake is attracted very quickly and efficiently to the desired feature.

show abstract

Section: A Contour Extraction Resultsmentioning

confidence: 99%

Section: B Level Set Representationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A geometric snake model for segmentation of medical imagery

Yezzi

Kichenassamy

Kumar

et al. 1997

IEEE Trans. Med. Imaging

487

299

View full text Add to dashboard Cite

show abstract

“…As it has been shown [21, 39], myocardial wall identification has become vital for calculating ejection-fraction ratio, heart output, ventricular volume ratio and muscle wall thickness, contributing to the identification of several medical pathologies. Then, as an example of a segmentation of this type of biological tissue, we show results of the segmentation of the left ventricle in human hearts in two and three dimensions.…”

Section: Numerical Implementationmentioning

confidence: 99%

Diffeomorphic Active Contours

Arrate¹,

Ratnanather²,

Younes³

2010

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

In this study we present a geometric flow approach to the segmentation of three-dimensional medical images obtained from magnetic resonance imaging (MRI) or computed tomography (CT) scan methods, by minimizing a cost function. This energy term is based on the intensity of the original image and its minimum is found following a gradient descent curve in an infinitedimensional space of diffeomorphisms (Diff) to preserve topology. The general framework is reminiscent of variational shape optimization methods, but remains closer to general developments on deformable template theory of geometric flows. In our case, the metric that provides the gradient is defined as a right invariant inner product on the tangent space ( at the identity of the group of diffeomorphisms, following the general Lie group approach suggested by Arnold [2]. To avoid local solutions of the optimization problem and to mitigate the influence of several sources of noise, a finite set of control points is defined on the boundary of the template binary images, yielding a projected gradient descent on Diff.

show abstract

“…al. (14) have performed other notable research using multi-step active deformation methods to describe ventricular wall segmentation. After identifying the outside heart wall, they improved the interior wall segmentation by using the information on the extraluminal boundary which is applied to the variational integration that is used to better control convergence of the interior wall.…”

Section: Introductionmentioning

confidence: 99%

Iterative active deformational methodology for tumor delineation: Evaluation across radiation treatment stage and volume

Shaffer

Thompson

et al. 2008

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: To introduce, implement, and assess an iterative modification to the active deformational image segmentation method as applied to cervical cancer tumors. Materials and Methods:A comparison by Jaccard similarity (J S ) between this active deformational method and manual segmentation was performed on tumors of various sizes across preradiation, 3 weeks postradiation, and 6 weeks postradiation using a General Linear Mixed Model across 121 studies from 52 patients with Stage IIB-IV cervical cancers. Results:The deformable segmentation method produced promising levels of agreement including J S factors of 0.71 Ϯ 0.11 in the preradiation studies. The analysis illustrated a rate of improvement in J S with increasing tumor volume that differed between the preradiation and 6 weeks postradiation stage (P ϭ 0.0474). In the large preradiated tumors each additional cm 3 of volume was associated with an increase or improvement in J S of 0.0008 (95% confidence interval [CI]: 0.0003, 0.0014). In the smaller postradiation tumors, each additional cm 3 of volume was associated with a more robust improvement in J S of 0.0046 (95% CI: 0.0009, 0.0082). Conclusion:Agreement was strongly affected by tumor volume, and its performance was most impacted across volume in the later stages of radiation therapy. The deformation-based segmentation method appears to demonstrate utility for delineating cervical cancer tumors, particularly in the earliest stages of radiation treatment, where agreement is greatest.

show abstract

Cardiac MR image segmentation using deformable models

Cited by 34 publications

References 3 publications

A geometric snake model for segmentation of medical imagery

A geometric snake model for segmentation of medical imagery

Diffeomorphic Active Contours

Iterative active deformational methodology for tumor delineation: Evaluation across radiation treatment stage and volume

Contact Info

Product

Resources

About