Space-time segmentation using level set active contours applied to myocardial gated SPECT

Abstract

This paper presents a new variational method for the segmentation of a moving object against a still background, over a sequence of (2-D or 3-D) image frames. The method is illustrated in application to myocardial gated SPECT data, and incorporates a level set framework to handle topological changes while providing closed boundaries. The key innovation is the introduction of a geometrical constraint into the derivation of the Euler-Lagrange equations, such that the segmentation of each individual frame can be interpreted as a closed boundary of an object (an isolevel of a set of hyper-surfaces) while integrating information over the entire sequence. This results in the definition of an evolution velocity normal to the object boundary. Applying this method to 3-D myocardial gated SPECT sequences, the left ventricle endocardial and epicardial limits can be computed in each frame. This space-time segmentation method was tested on simulated and clinical 3-D myocardial gated SPECT sequences and the corresponding ejection fractions were computed.

A dynamic elastic model for segmentation and tracking of the heart in MR image sequences

“…Lorenzo-Valdes et al (2004) proposed using a 4D probabilistic atlas together with the expectation maximization algorithm to perform the segmentation of Cine MRI. Debreuve et al (2001) proposed a space-time level-set method for the segmentation of myocardial SPECT images, taking the whole sequence as one 4D image. More recently, methods derived from mathematical morphology have appeared: Cousty et al presented a framework based on discrete watersheds, also working in 4D (Cousty et al, 2010).…”

A Semi-Automatic Clustering-Based Level Set Method for Segmentation of Endocardium from MSCT Images

“…Other similar approaches using shape information with level sets include [3], [4] and [5]. In [6], Debreuve et al introduced a level set method for segmenting the myocardial from SPECT images. An energy function of the level set equation was minimized by assuming a constant intensity approximation of the object and background regions.…”

Automatic quantification of three-dimensional kinetic energy in gated myocardial perfusion single-photon-emission computerized tomography improved by a multiresolution technique

“…In cardiac SPECT images using a radionuclide such as 201 TI and 99m Tc-MIBI, the principal structure is the myocardium, although it is a reasonable assumption that the total intensity of voxels within the ventricle is conserved between successive frames, and that the motion of connected tissue within the myocardium should be smooth. 21,22 Equation ͑1͒ is usually called the OF constraint ͑OFC͒ and to evaluate the flow velocity, Horn and Schunck 20 introduced a ''smoothness'' constraint in addition to the fundamental constraint. Using the Horn and Schunck method and the extension to three dimensions, the velocity flow field can be obtained by minimizing the functional:…”