Automatic Left Atrium Segmentation by Cutting the Blood Pool at Narrowings

Abstract

Abstract. This paper presents a method to extract heart structures from CTA and MRA data sets, in particular the left atrium. First, the segmented blood pool is subdivided at narrowings in small components. Second, these basic components are merged automatically so that they represent the different heart structures. The resulting cutting surfaces have a relatively small diameter compared to the diameter of the neighboring heart chambers. Both steps are controlled by only one fixed parameter. The method is fast and allows interactive postprocessing by the user. Experiments on various data sets show the accuracy, robustness and repeatability of this approach.

Automatic Segmentation of Left Atrial Geometry from Contrast-Enhanced Magnetic Resonance Images Using a Probabilistic Atlas

“…These junctions were exploited in [1,4] by making the appropriate cuts and separating the atrium from the rest of the image. For finding narrow junctions within a binary image, a Voronoi Tessellation of the Euclidean Distance Transform (EDT) image is computed.…”

“…By definition, each Voronoi cell only comprise of voxels that are closest to its centre. As in [1], each Voronoi cell has a size and the separating surface between cells are also given a size. The separating surface are shared borders with a neighbouring cell and illustrated in Fig.…”

“…where D(·) is the diameter function of a Voronoi cell and S is the separating surface or interface between two adjacent cells. Following [1] and [4], we use D(·) as the EDT value of the cell and surface centres. Larger values of η indicate a Voronoi neighbourhood with a narrowing.…”

“…Contrast-enhanced Magnetic Resonance Angiography provides good quality imaging data for segmenting the atrial anatomy. Previous work on the segmentation of the left atrium include segmentation from MRA/CTA [1,4], CT [7] and X-ray angiography [8]. Our proposed method adopts the Voronoi framework of [1,4].…”

“…Previous work on the segmentation of the left atrium include segmentation from MRA/CTA [1,4], CT [7] and X-ray angiography [8]. Our proposed method adopts the Voronoi framework of [1,4]. However, the segmentation method is based on the minimization of an energy functional with a spatial and intensity prior.…”