Landmark optimization using local curvature for point-based nonlinear rodent brain image registration

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Purpose. To develop a technique to automate landmark selection for point-based interpolating transformations for nonlinear medical image registration. Materials and Methods. Interpolating transformations were calculated from homologous point landmarks on the source (image to be transformed) and target (reference image). Point landmarks are placed at regular intervals on contours of anatomical features, and their positions are optimized along the contour surface by a function composed of curvature similarity and displacements of the homologous landmarks. The method was evaluated in two cases (n = 5 each). In one, MRI was registered to histological sections; in the second, geometric distortions in EPI MRI were corrected. Normalized mutual information and target registration error were calculated to compare the registration accuracy of the automatically and manually generated landmarks. Results. Statistical analyses demonstrated significant improvement (P < 0.05) in registration accuracy by landmark optimization in most data sets and trends towards improvement (P < 0.1) in others as compared to manual landmark selection.

Original languageEnglish (US)
Article number635207
JournalInternational Journal of Biomedical Imaging
Volume2012
DOIs
StatePublished - Jan 1 2012

Fingerprint

Rodentia
Brain
Datasets

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{071a72568f6c4f5c8b7bfeefa1d02b67,
title = "Landmark optimization using local curvature for point-based nonlinear rodent brain image registration",
abstract = "Purpose. To develop a technique to automate landmark selection for point-based interpolating transformations for nonlinear medical image registration. Materials and Methods. Interpolating transformations were calculated from homologous point landmarks on the source (image to be transformed) and target (reference image). Point landmarks are placed at regular intervals on contours of anatomical features, and their positions are optimized along the contour surface by a function composed of curvature similarity and displacements of the homologous landmarks. The method was evaluated in two cases (n = 5 each). In one, MRI was registered to histological sections; in the second, geometric distortions in EPI MRI were corrected. Normalized mutual information and target registration error were calculated to compare the registration accuracy of the automatically and manually generated landmarks. Results. Statistical analyses demonstrated significant improvement (P < 0.05) in registration accuracy by landmark optimization in most data sets and trends towards improvement (P < 0.1) in others as compared to manual landmark selection.",
author = "Yutong Liu and Sajja, {Balasrinivasa R} and Uberti, {Mariano G.} and Gendelman, {Howard Eliot} and Kielian, {Tammy L} and Boska, {Michael D.}",
year = "2012",
month = "1",
day = "1",
doi = "10.1155/2012/635207",
language = "English (US)",
volume = "2012",
journal = "International Journal of Biomedical Imaging",
issn = "1687-4188",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Landmark optimization using local curvature for point-based nonlinear rodent brain image registration

AU - Liu, Yutong

AU - Sajja, Balasrinivasa R

AU - Uberti, Mariano G.

AU - Gendelman, Howard Eliot

AU - Kielian, Tammy L

AU - Boska, Michael D.

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Purpose. To develop a technique to automate landmark selection for point-based interpolating transformations for nonlinear medical image registration. Materials and Methods. Interpolating transformations were calculated from homologous point landmarks on the source (image to be transformed) and target (reference image). Point landmarks are placed at regular intervals on contours of anatomical features, and their positions are optimized along the contour surface by a function composed of curvature similarity and displacements of the homologous landmarks. The method was evaluated in two cases (n = 5 each). In one, MRI was registered to histological sections; in the second, geometric distortions in EPI MRI were corrected. Normalized mutual information and target registration error were calculated to compare the registration accuracy of the automatically and manually generated landmarks. Results. Statistical analyses demonstrated significant improvement (P < 0.05) in registration accuracy by landmark optimization in most data sets and trends towards improvement (P < 0.1) in others as compared to manual landmark selection.

AB - Purpose. To develop a technique to automate landmark selection for point-based interpolating transformations for nonlinear medical image registration. Materials and Methods. Interpolating transformations were calculated from homologous point landmarks on the source (image to be transformed) and target (reference image). Point landmarks are placed at regular intervals on contours of anatomical features, and their positions are optimized along the contour surface by a function composed of curvature similarity and displacements of the homologous landmarks. The method was evaluated in two cases (n = 5 each). In one, MRI was registered to histological sections; in the second, geometric distortions in EPI MRI were corrected. Normalized mutual information and target registration error were calculated to compare the registration accuracy of the automatically and manually generated landmarks. Results. Statistical analyses demonstrated significant improvement (P < 0.05) in registration accuracy by landmark optimization in most data sets and trends towards improvement (P < 0.1) in others as compared to manual landmark selection.

UR - http://www.scopus.com/inward/record.url?scp=81555225324&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81555225324&partnerID=8YFLogxK

U2 - 10.1155/2012/635207

DO - 10.1155/2012/635207

M3 - Article

VL - 2012

JO - International Journal of Biomedical Imaging

JF - International Journal of Biomedical Imaging

SN - 1687-4188

M1 - 635207

ER -