DaWachinger

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

Three-dimensional Ultrasound Mosaicing

Diploma thesis

Three-Dimensional Ultrasound Mosaicing

Thesis by: Christian Wachinger
Advisor: Prof. Nassir Navab, Frank Sauer, Ph.D.
Supervision by: Wolfgang Wein, Ali Khamene?
in collaboration with Siemens Corporate Research, Princeton, USA
Due date: March 2007

Awarded with the Werner von Siemens Excellence Award 2007.

Abstract

The creation of 2D ultrasound mosaics is becoming a common clinical practice with a high clinical value. The next step coming along with the increasing availability of 2D array transducers is the creation of 3D mosaics. In the literature of ultrasound registration, the alignment of multiple images has not yet been addressed. Therefore, we propose registration strategies, which are able to cope with problems arising by multiple image alignment. We use pair-wise registration with a consecutive Lie normalization and simultaneous registration, which urges the usage of multivariate similarity measures. In this thesis, we propose alternative multivariate extensions based on a maximum likelihood framework. Due to the higher computational cost of simultaneous registration, we describe possibilities for speeding them up, among others, the usage of a stochastic pyramid.

We also present methods to reduce speckle noise and to detect shadow in ultrasonographic volumes, to improve the overall registration performance. For the compounding of the final volume, a variety of approaches are listed, ranging from general to advanced ones. Experimental results, on multiple ultrasound data sets, show the good performance of the proposed registration strategies and similarity measures.

keywords: ultrasound registration, three-dimensional (3D) mosaicing, ultrasound stitching, ultrasound compounding, extended field-of-view ultrasound, simultaneous registration, lie normalization, pairwise registration, multivariate similarity measures, global registration

Videos

  • 3D Baby Mosaicing
    • Shaking Baby: Demonstration of simultaneous registration with a newly developed similarity measure on baby phantom data

  • Cylinder Study: A simplified model of the human legs on ultrasoud images is a cylinder. Each acquisition shows a plate; scanning around the leg and fusing all the scans should result in a cylinder. Using pairwise registration can first lead to an accumulation of registration errors, resulting in a spiral instead of an cylinder. And second, to a total overlap of the images. In contrast, simultaneous registration leads to a stable and robust registration of the plates.

  • Triangle Study: Having three times the same volume consisting of two cylinders each. Starting from a rough initial alignment, volume number 1 is rotationally displaced by a certain degree. First, a pairwise registration is tried, leading to a bad final result. Second, a simultaneous registration leads to the desired triangular result.




Images

  • Vessel:

  • Baby:




Ressources:

Publications:

C. Wachinger, N. Navab
A Contextual Maximum Likelihood Framework for Modeling Image Registration
IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Providence, Rhode Island (USA), June 2012. (bib)
C. Wachinger, N. Navab
Manifold Learning for Multi-Modal Image Registration
21st British Machine Vision Conference (BMVC), Aberystwyth, United Kingdom, August 31 - September 3 2010. (bib)
C. Wachinger, N. Navab
Structural Image Representation for Image Registration
IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, California, June 2010. Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA). (bib)
C. Wachinger, N. Navab
Alignment of Viewing-Angle Dependent Ultrasound Images.
Medical Image Computing and Computer-Assisted Intervention (MICCAI), London, UK, September 20-24 2009 (bib)
O. Kutter, W. Wein, N. Navab
Multi-modal registration based Ultrasound Mosaicing
Medical Image Computing and Computer-Assisted Intervention (MICCAI), London, UK, September 20-24 2009 (bib)
C. Wachinger, N. Navab
Similarity Metrics and Efficient Optimization for Simultaneous Registration
IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Miami, Florida (USA), June 2009. (bib)
C. Wachinger, B. Glocker, J. Zeltner, N. Paragios, N. Komodakis, M. S. Hansen, N. Navab
Deformable Mosaicing for Whole-body MRI
Medical Image Computing and Computer-Assisted Intervention (MICCAI), New York, USA, September 6-10 2008 (bib)
C. Wachinger
Insights Into Three-Dimensional Ultrasound Mosaicing: Strategies for the alignment of multiple ultrasound volumes
VDM Verlag Dr. Müller, Stuttgart, ISBN: 3836488264, December 23 2008, Amazon (bib)
C. Wachinger, W. Wein, N. Navab
Registration Strategies and Similarity Measures for Three-dimensional Ultrasound Mosaicing
Journal of the University Radiologists, Academic Radiology 15 (11), pp.1404-1415. Special issue of MICCAI 2007 (bib)
S. Atasoy, D. Noonan, S. Benhimane, N. Navab, G. Z. Yang
A Global Approach for Automatic Fibroscopic Video Mosaicing in Minimally Invasive Diagnosis
Medical Image Computing and Computer-Assisted Intervention, MICCAI, 2008, New York, USA, September 6-10 2008 (bib)
C. Wachinger, R. Shams, N. Navab
Estimation of Acoustic Impedance from Multiple Ultrasound Images with Application to Spatial Compounding
IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Anchorage, Alaska (USA), June 2008. Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA). (bib)
C. Wachinger, N. Navab
Ultrasound Specific Similarity Measures for Three-Dimensional Mosaicing.
SPIE Medical Imaging, San Diego, California, USA, 16-21 February 2008 (bib)
C. Wachinger, W. Wein, N. Navab
Three-Dimensional Ultrasound Mosaicing
Medical Image Computing and Computer-Assisted Intervention, MICCAI, 2007, Brisbane, Australia, October 29 - November 2 2007 (bib)
C. Wachinger, W. Wein, N. Navab
Ultrasound Volume Stitching
6. Jahrestagung der Deutschen Gesellschaft für Computer-und Roboter-Assistierte Chirurgie (CURAC 2007), Karlsruhe, Germany, October 2007 (bib)



Students.ProjectForm
Title: Three-Dimensional Ultrasound Mosaicing
Abstract: The creation of 2D ultrasound mosaics is becoming a common clinical practice with a high clinical value. The next step coming along with the increasing availability of 2D array transducers is the creation of 3D mosaics. In the literature of ultrasound registration, the alignment of multiple images has not yet been addressed. Therefore, we propose registration strategies, which are able to cope with problems arising by multiple image alignment. We use pair-wise registration with a consecutive Lie normalization and simultaneous registration, which urges the usage of multivariate similarity measures. In this thesis, we propose alternative multivariate extensions based on a maximum likelihood framework. Due to the higher computational cost of simultaneous registration, we describe possibilities for speeding them up, among others, the usage of a stochastic pyramid. We also present methods to reduce speckle noise and to detect shadow in ultrasonographic volumes, to improve the overall registration performance. For the compounding of the final volume, a variety of approaches are listed, ranging from general to advanced ones. Experimental results, on multiple ultrasound data sets, show the good performance of the proposed registration strategies and similarity measures.
Student: Christian Wachinger
Director: Nassir Navab
Supervisor: Wolfgang Wein
Type: Diploma Thesis
Area:  
Status: finished
Start: 2006/10/01
Finish: 2007/03/30


Edit | Attach | Refresh | Diffs | More | Revision r1.25 - 09 Dec 2007 - 16:15 - ChristianWachinger