HAMMER: Hierarchical attribute matching mechanism for elastic registration

Dinggang Shen, Christos Davatzikos

Research output: Contribution to journalArticle

811 Citations (Scopus)

Abstract

A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate very high accuracy in superposition of images from different subjects. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e., a set of geometric moment invariants (GMIs) that are defined on each voxel in an image and are calculated from the tissue maps, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure; it also helps reduce local minima, by reducing ambiguity in potential matches. This is a fundamental deviation of our method, referred to as the hierarchical attribute matching mechanism for elastic registration (HAMMER), from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e., suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional smooth energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting the driving features that have distinct attribute vectors, thus, drastically reducing ambiguity in finding correspondence. A number of experiments demonstrate that the proposed algorithm results in accurate superposition of image data from individuals with significant anatomical differences.

Original languageEnglish
Pages (from-to)1421-1439
Number of pages19
JournalIEEE Transactions on Medical Imaging
Volume21
Issue number11
DOIs
Publication statusPublished - 2002 Nov 1
Externally publishedYes

Fingerprint

Magnetic resonance
Brain
Anatomy
Magnetic Resonance Spectroscopy
Tissue
Experiments

Keywords

  • Attribute vectors
  • Average brain
  • Deformable registration
  • Geometric moment invariants
  • Hierarchical deformation mechanism
  • Multigrid formulation
  • Statistical atlases

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

HAMMER : Hierarchical attribute matching mechanism for elastic registration. / Shen, Dinggang; Davatzikos, Christos.

In: IEEE Transactions on Medical Imaging, Vol. 21, No. 11, 01.11.2002, p. 1421-1439.

Research output: Contribution to journalArticle

@article{1130a1096dea4fec912040cdac1de746,
title = "HAMMER: Hierarchical attribute matching mechanism for elastic registration",
abstract = "A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate very high accuracy in superposition of images from different subjects. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e., a set of geometric moment invariants (GMIs) that are defined on each voxel in an image and are calculated from the tissue maps, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure; it also helps reduce local minima, by reducing ambiguity in potential matches. This is a fundamental deviation of our method, referred to as the hierarchical attribute matching mechanism for elastic registration (HAMMER), from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e., suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional smooth energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting the driving features that have distinct attribute vectors, thus, drastically reducing ambiguity in finding correspondence. A number of experiments demonstrate that the proposed algorithm results in accurate superposition of image data from individuals with significant anatomical differences.",
keywords = "Attribute vectors, Average brain, Deformable registration, Geometric moment invariants, Hierarchical deformation mechanism, Multigrid formulation, Statistical atlases",
author = "Dinggang Shen and Christos Davatzikos",
year = "2002",
month = "11",
day = "1",
doi = "10.1109/TMI.2002.803111",
language = "English",
volume = "21",
pages = "1421--1439",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "11",

}

TY - JOUR

T1 - HAMMER

T2 - Hierarchical attribute matching mechanism for elastic registration

AU - Shen, Dinggang

AU - Davatzikos, Christos

PY - 2002/11/1

Y1 - 2002/11/1

N2 - A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate very high accuracy in superposition of images from different subjects. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e., a set of geometric moment invariants (GMIs) that are defined on each voxel in an image and are calculated from the tissue maps, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure; it also helps reduce local minima, by reducing ambiguity in potential matches. This is a fundamental deviation of our method, referred to as the hierarchical attribute matching mechanism for elastic registration (HAMMER), from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e., suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional smooth energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting the driving features that have distinct attribute vectors, thus, drastically reducing ambiguity in finding correspondence. A number of experiments demonstrate that the proposed algorithm results in accurate superposition of image data from individuals with significant anatomical differences.

AB - A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate very high accuracy in superposition of images from different subjects. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e., a set of geometric moment invariants (GMIs) that are defined on each voxel in an image and are calculated from the tissue maps, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure; it also helps reduce local minima, by reducing ambiguity in potential matches. This is a fundamental deviation of our method, referred to as the hierarchical attribute matching mechanism for elastic registration (HAMMER), from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e., suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional smooth energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting the driving features that have distinct attribute vectors, thus, drastically reducing ambiguity in finding correspondence. A number of experiments demonstrate that the proposed algorithm results in accurate superposition of image data from individuals with significant anatomical differences.

KW - Attribute vectors

KW - Average brain

KW - Deformable registration

KW - Geometric moment invariants

KW - Hierarchical deformation mechanism

KW - Multigrid formulation

KW - Statistical atlases

UR - http://www.scopus.com/inward/record.url?scp=0036880516&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036880516&partnerID=8YFLogxK

U2 - 10.1109/TMI.2002.803111

DO - 10.1109/TMI.2002.803111

M3 - Article

C2 - 12575879

AN - SCOPUS:0036880516

VL - 21

SP - 1421

EP - 1439

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 11

ER -