SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images

Martin Urschler, Joachim Bauer, Hendrik Ditt, Horst Bischof

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Abstract

Nonlinear image registration is a prerequisite for various medical image analysis applications. Many data acquisition protocols suffer from problems due to breathing motion which has to be taken into account for further analysis. Intensity based nonlinear registration is often used to align differing images, however this requires a large computational effort, is sensitive to intensity variations and has problems with matching small structures. In this work a feature-based image registration method is proposed that combines runtime efficiency with good registration accuracy by making use of a fully automatic feature matching and registration approach. The algorithm stages are 3D corner detection, calculation of local (SIFT) and global (Shape Context) 3D descriptors, robust feature matching and calculation of a dense displacement field. An evaluation of the algorithm on seven synthetic and four clinical data sets is presented. The quantitative and qualitative evaluations show lower runtime and superior results when compared to the Demons algorithm.
Original languageEnglish
Title of host publicationComputer Vision Approaches to Medical Image Analysis
Subtitle of host publicationSecond International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers
EditorsReinhard R. Beichel, Milan Sonka
Place of PublicationBerlin Heidelberg
PublisherSpringer
Pages73-84
Volume4241
ISBN (Electronic)978-3-540-46258-3
ISBN (Print)978-3-540-46257-6
DOIs
Publication statusPublished - 2006

Publication series

NameLecture Notes in Computer Science

Fingerprint

Image registration
Image analysis
Data acquisition

Fields of Expertise

  • Information, Communication & Computing

Cite this

Urschler, M., Bauer, J., Ditt, H., & Bischof, H. (2006). SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images. In R. R. Beichel, & M. Sonka (Eds.), Computer Vision Approaches to Medical Image Analysis: Second International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers (Vol. 4241, pp. 73-84). (Lecture Notes in Computer Science). Berlin Heidelberg: Springer. https://doi.org/10.1007/11889762_7

SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images. / Urschler, Martin; Bauer, Joachim; Ditt, Hendrik; Bischof, Horst.

Computer Vision Approaches to Medical Image Analysis: Second International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers. ed. / Reinhard R. Beichel; Milan Sonka. Vol. 4241 Berlin Heidelberg : Springer, 2006. p. 73-84 (Lecture Notes in Computer Science).

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Urschler, M, Bauer, J, Ditt, H & Bischof, H 2006, SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images. in RR Beichel & M Sonka (eds), Computer Vision Approaches to Medical Image Analysis: Second International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers. vol. 4241, Lecture Notes in Computer Science, Springer, Berlin Heidelberg, pp. 73-84. https://doi.org/10.1007/11889762_7
Urschler M, Bauer J, Ditt H, Bischof H. SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images. In Beichel RR, Sonka M, editors, Computer Vision Approaches to Medical Image Analysis: Second International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers. Vol. 4241. Berlin Heidelberg: Springer. 2006. p. 73-84. (Lecture Notes in Computer Science). https://doi.org/10.1007/11889762_7
Urschler, Martin ; Bauer, Joachim ; Ditt, Hendrik ; Bischof, Horst. / SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images. Computer Vision Approaches to Medical Image Analysis: Second International ECCV Workshop, CVAMIA 2006 Graz, Austria, May 12, 2006 Revised Papers. editor / Reinhard R. Beichel ; Milan Sonka. Vol. 4241 Berlin Heidelberg : Springer, 2006. pp. 73-84 (Lecture Notes in Computer Science).
@inbook{1ea8ae1cb3b24fd0b31afe15d078c98b,
title = "SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images",
abstract = "Nonlinear image registration is a prerequisite for various medical image analysis applications. Many data acquisition protocols suffer from problems due to breathing motion which has to be taken into account for further analysis. Intensity based nonlinear registration is often used to align differing images, however this requires a large computational effort, is sensitive to intensity variations and has problems with matching small structures. In this work a feature-based image registration method is proposed that combines runtime efficiency with good registration accuracy by making use of a fully automatic feature matching and registration approach. The algorithm stages are 3D corner detection, calculation of local (SIFT) and global (Shape Context) 3D descriptors, robust feature matching and calculation of a dense displacement field. An evaluation of the algorithm on seven synthetic and four clinical data sets is presented. The quantitative and qualitative evaluations show lower runtime and superior results when compared to the Demons algorithm.",
author = "Martin Urschler and Joachim Bauer and Hendrik Ditt and Horst Bischof",
year = "2006",
doi = "10.1007/11889762_7",
language = "English",
isbn = "978-3-540-46257-6",
volume = "4241",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "73--84",
editor = "Beichel, {Reinhard R.} and Milan Sonka",
booktitle = "Computer Vision Approaches to Medical Image Analysis",

}

TY - CHAP

T1 - SIFT and Shape Context for Feature-Based Nonlinear Registration of Thoracic CT Images

AU - Urschler, Martin

AU - Bauer, Joachim

AU - Ditt, Hendrik

AU - Bischof, Horst

PY - 2006

Y1 - 2006

N2 - Nonlinear image registration is a prerequisite for various medical image analysis applications. Many data acquisition protocols suffer from problems due to breathing motion which has to be taken into account for further analysis. Intensity based nonlinear registration is often used to align differing images, however this requires a large computational effort, is sensitive to intensity variations and has problems with matching small structures. In this work a feature-based image registration method is proposed that combines runtime efficiency with good registration accuracy by making use of a fully automatic feature matching and registration approach. The algorithm stages are 3D corner detection, calculation of local (SIFT) and global (Shape Context) 3D descriptors, robust feature matching and calculation of a dense displacement field. An evaluation of the algorithm on seven synthetic and four clinical data sets is presented. The quantitative and qualitative evaluations show lower runtime and superior results when compared to the Demons algorithm.

AB - Nonlinear image registration is a prerequisite for various medical image analysis applications. Many data acquisition protocols suffer from problems due to breathing motion which has to be taken into account for further analysis. Intensity based nonlinear registration is often used to align differing images, however this requires a large computational effort, is sensitive to intensity variations and has problems with matching small structures. In this work a feature-based image registration method is proposed that combines runtime efficiency with good registration accuracy by making use of a fully automatic feature matching and registration approach. The algorithm stages are 3D corner detection, calculation of local (SIFT) and global (Shape Context) 3D descriptors, robust feature matching and calculation of a dense displacement field. An evaluation of the algorithm on seven synthetic and four clinical data sets is presented. The quantitative and qualitative evaluations show lower runtime and superior results when compared to the Demons algorithm.

U2 - 10.1007/11889762_7

DO - 10.1007/11889762_7

M3 - Chapter

SN - 978-3-540-46257-6

VL - 4241

T3 - Lecture Notes in Computer Science

SP - 73

EP - 84

BT - Computer Vision Approaches to Medical Image Analysis

A2 - Beichel, Reinhard R.

A2 - Sonka, Milan

PB - Springer

CY - Berlin Heidelberg

ER -