Estimating myocardial motion by 4D image warping

Hari Sundar, Harold Litt, Dinggang Shen

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

A method for spatio-temporally smooth and consistent estimation of cardiac motion from MR cine sequences is proposed. Myocardial motion is estimated within a four-dimensional (4D) registration framework, in which all three-dimensional (3D) images obtained at different cardiac phases are simultaneously registered. This facilitates spatio-temporally consistent estimation of motion as opposed to other registration-based algorithms which estimate the motion by sequentially registering one frame to another. To facilitate image matching, an attribute vector (AV) is constructed for each point in the image, and is intended to serve as a "morphological signature" of that point. The AV includes intensity, boundary, and geometric moment invariants (GMIs). Hierarchical registration of two image sequences is achieved by using the most distinctive points for initial registration of two sequences and gradually adding less-distinctive points to refine the registration. Experimental results on real data demonstrate good performance of the proposed method for cardiac image registration and motion estimation. The motion estimation is validated via comparisons with motion estimates obtained from MR images with myocardial tagging.

Original languageEnglish
Pages (from-to)2514-2526
Number of pages13
JournalPattern Recognition
Volume42
Issue number11
DOIs
Publication statusPublished - 2009 Nov 1
Externally publishedYes

Fingerprint

Motion estimation
Image matching
Image registration

Keywords

  • Cardiac motion estimation
  • Image registration
  • Spatio-temporal normalization

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Estimating myocardial motion by 4D image warping. / Sundar, Hari; Litt, Harold; Shen, Dinggang.

In: Pattern Recognition, Vol. 42, No. 11, 01.11.2009, p. 2514-2526.

Research output: Contribution to journalArticle

Sundar, Hari ; Litt, Harold ; Shen, Dinggang. / Estimating myocardial motion by 4D image warping. In: Pattern Recognition. 2009 ; Vol. 42, No. 11. pp. 2514-2526.
@article{b4c9272971d54a70810d09e86266359e,
title = "Estimating myocardial motion by 4D image warping",
abstract = "A method for spatio-temporally smooth and consistent estimation of cardiac motion from MR cine sequences is proposed. Myocardial motion is estimated within a four-dimensional (4D) registration framework, in which all three-dimensional (3D) images obtained at different cardiac phases are simultaneously registered. This facilitates spatio-temporally consistent estimation of motion as opposed to other registration-based algorithms which estimate the motion by sequentially registering one frame to another. To facilitate image matching, an attribute vector (AV) is constructed for each point in the image, and is intended to serve as a {"}morphological signature{"} of that point. The AV includes intensity, boundary, and geometric moment invariants (GMIs). Hierarchical registration of two image sequences is achieved by using the most distinctive points for initial registration of two sequences and gradually adding less-distinctive points to refine the registration. Experimental results on real data demonstrate good performance of the proposed method for cardiac image registration and motion estimation. The motion estimation is validated via comparisons with motion estimates obtained from MR images with myocardial tagging.",
keywords = "Cardiac motion estimation, Image registration, Spatio-temporal normalization",
author = "Hari Sundar and Harold Litt and Dinggang Shen",
year = "2009",
month = "11",
day = "1",
doi = "10.1016/j.patcog.2009.04.022",
language = "English",
volume = "42",
pages = "2514--2526",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "11",

}

TY - JOUR

T1 - Estimating myocardial motion by 4D image warping

AU - Sundar, Hari

AU - Litt, Harold

AU - Shen, Dinggang

PY - 2009/11/1

Y1 - 2009/11/1

N2 - A method for spatio-temporally smooth and consistent estimation of cardiac motion from MR cine sequences is proposed. Myocardial motion is estimated within a four-dimensional (4D) registration framework, in which all three-dimensional (3D) images obtained at different cardiac phases are simultaneously registered. This facilitates spatio-temporally consistent estimation of motion as opposed to other registration-based algorithms which estimate the motion by sequentially registering one frame to another. To facilitate image matching, an attribute vector (AV) is constructed for each point in the image, and is intended to serve as a "morphological signature" of that point. The AV includes intensity, boundary, and geometric moment invariants (GMIs). Hierarchical registration of two image sequences is achieved by using the most distinctive points for initial registration of two sequences and gradually adding less-distinctive points to refine the registration. Experimental results on real data demonstrate good performance of the proposed method for cardiac image registration and motion estimation. The motion estimation is validated via comparisons with motion estimates obtained from MR images with myocardial tagging.

AB - A method for spatio-temporally smooth and consistent estimation of cardiac motion from MR cine sequences is proposed. Myocardial motion is estimated within a four-dimensional (4D) registration framework, in which all three-dimensional (3D) images obtained at different cardiac phases are simultaneously registered. This facilitates spatio-temporally consistent estimation of motion as opposed to other registration-based algorithms which estimate the motion by sequentially registering one frame to another. To facilitate image matching, an attribute vector (AV) is constructed for each point in the image, and is intended to serve as a "morphological signature" of that point. The AV includes intensity, boundary, and geometric moment invariants (GMIs). Hierarchical registration of two image sequences is achieved by using the most distinctive points for initial registration of two sequences and gradually adding less-distinctive points to refine the registration. Experimental results on real data demonstrate good performance of the proposed method for cardiac image registration and motion estimation. The motion estimation is validated via comparisons with motion estimates obtained from MR images with myocardial tagging.

KW - Cardiac motion estimation

KW - Image registration

KW - Spatio-temporal normalization

UR - http://www.scopus.com/inward/record.url?scp=67649395501&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649395501&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2009.04.022

DO - 10.1016/j.patcog.2009.04.022

M3 - Article

AN - SCOPUS:67649395501

VL - 42

SP - 2514

EP - 2526

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 11

ER -