Matching of anatomical tree structures for registration of medical images

Metzen, Jan Hendrik; Kröger, Tim; Schenk, Andrea; Zidowitz, Stephan; Peitgen, Heinz‐Otto; Jiang, Xiaoyi

doi:10.1016/j.imavis.2008.04.002

Cited by 26 publications

(22 citation statements)

References 31 publications

(57 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A good overview of early methods is also available in (Graham and Higgins, 2006a,b) and (Metzen et al, 2007(Metzen et al, , 2009, where the authors explained, among others, why searching a graph isomorphism is not directly applicable to match anatomical trees. In (Pisupati et al, 1996a,b) a matching algorithm based on dynamic programming has been used to find the correspondences between two airway trees, A and B, representing airway skeletons from low and high pressures, respectively.…”

Section: Related Workmentioning

confidence: 99%

“…Because finding the maximum clique in an association graph is an NP-complete problem, Tschirren et al (2005) and Metzen et al (2007Metzen et al ( , 2009 have proposed heuristics to make it computationally tractable. Processing times reported for portal-vein trees and bronchial trees containing up to two hundred nodes were between three and six minutes, and 84 to 95% of the matches in bronchial trees were correct.…”

Section: Related Workmentioning

confidence: 99%

“…path-to-path distances extracted from point-to-point branch information (Metzen et al, 2007(Metzen et al, , 2009, branch or path lengths (Graham and Higgins, 2006a;Kaftan et al, 2006;Smeets et al, 2010), branch orientation (Pisupati et al, 1996b;Tschirren et al, 2005;Kaftan et al, 2006;Graham and Higgins, 2006a), branch diameter (Pisupati et al, 1996b), Euclidean distance between initial and final nodes of paths (Tschirren et al, 2005;Pisupati et al, 1996b;Graham and Higgins, 2006a), 3D context and local statistical moments of point distributions (Bülow et al, 2006), and n-SIFT descriptor (Smeets et al, 2010). As for the topology information, it was managed by defining a set of valid matches, based on their topological differences (Graham and Higgins, 2006a,b) or by restricting the comparison of paths based on their topological relations.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

“…As for the topology information, it was managed by defining a set of valid matches, based on their topological differences (Graham and Higgins, 2006a,b) or by restricting the comparison of paths based on their topological relations. The restriction of path comparison was used for the creation of association graphs, as in (Tschirren et al, 2005;Metzen et al, 2007Metzen et al, , 2009, or for the evaluation of possible matches (Pisupati et al, 1996b). In the tree-space introduced by Feragen et al (2015), topology changes were managed by evaluating the geodesic distance across different quadrants, whereas the geometric information involved in this distance corresponded to the branch lengths.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

See 3 more Smart Citations

A tree-matching algorithm: Application to airways in CT images of subjects with the acute respiratory distress syndrome

Pinzón

Hoyos

Richard

et al. 2017

Medical Image Analysis

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Accepted Manuscriptmentioning

confidence: 99%

Section: Accepted Manuscriptmentioning

confidence: 99%

See 2 more Smart Citations

A tree-matching algorithm: Application to airways in CT images of subjects with the acute respiratory distress syndrome

Pinzón

Hoyos

Richard

et al. 2017

Medical Image Analysis

View full text Add to dashboard Cite

“…Tree-structures appear, for instance, as airway trees in the lungs [20,21], as blood vessel trees [13], or as skeleta of more general shapes [4,9,10,15,17,19]. Anatomical and biological trees carry information about the organ or organism that contains them, and many pattern recognition algorithms, e.g., in computer vision and medical image analysis, require a distance measure between treestructures as input [5,10,15].…”

Section: Introductionmentioning

confidence: 99%

Complexity of Computing Distances between Geometric Trees

Feragen

2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Geometric trees can be formalized as unordered combinatorial trees whose edges are endowed with geometric information. Examples are skeleta of shapes from images; anatomical tree-structures such as blood vessels; or phylogenetic trees. An inter-tree distance measure is a basic prerequisite for many pattern recognition and machine learning methods to work on anatomical, phylogenetic or skeletal trees. Standard distance measures between trees, such as tree edit distance, can be readily translated to the geometric tree setting. It is well-known that the tree edit distance for unordered trees is generally NP complete to compute. However, the classical proof of NP completeness depends on a particular case of edit distance with integer edit costs for trees with discrete labels, and does not obviously carry over to the class of geometric trees. The reason is that edge geometry is encoded in continuous scalar or vector attributes, allowing for continuous edit paths from one tree to another, rather than finite, discrete edit sequences with discrete costs for discrete label sets. In this paper, we explain why the proof does not carry over directly to the continuous setting, and why it does not work for the important class of trees with scalar-valued edge attributes, such as edge length. We prove the NP completeness of tree edit distance and another natural distance measure, QED, for geometric trees with vector valued edge attributes.

show abstract

Methods of graph network reconstruction in personalized medicine

Danilov

Ivanov

Pryamonosov

et al. 2015

Numer Methods Biomed Eng

View full text Add to dashboard Cite

The paper addresses methods for generation of individualized computational domains on the basis of medical imaging dataset. The computational domains will be used in one-dimensional (1D) and three-dimensional (3D)-1D coupled hemodynamic models. A 1D hemodynamic model employs a 1D network of a patient-specific vascular network with large number of vessels. The 1D network is the graph with nodes in the 3D space which bears additional geometric data such as length and radius of vessels. A 3D hemodynamic model requires a detailed 3D reconstruction of local parts of the vascular network. We propose algorithms which extend the automated segmentation of vascular and tubular structures, generation of centerlines, 1D network reconstruction, correction, and local adaptation. We consider two modes of centerline representation: (i) skeletal segments or sets of connected voxels and (ii) curved paths with corresponding radii. Individualized reconstruction of 1D networks depends on the mode of centerline representation. Efficiency of the proposed algorithms is demonstrated on several examples of 1D network reconstruction. The networks can be used in modeling of blood flows as well as other physiological processes in tubular structures. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Matching of anatomical tree structures for registration of medical images

Cited by 26 publications

References 31 publications

A tree-matching algorithm: Application to airways in CT images of subjects with the acute respiratory distress syndrome

A tree-matching algorithm: Application to airways in CT images of subjects with the acute respiratory distress syndrome

Complexity of Computing Distances between Geometric Trees

Methods of graph network reconstruction in personalized medicine

Contact Info

Product

Resources

About