A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields

Jan Rühaak; Lars König; Marc Hallmann; Nils Papenberg; Stefan Heldmann; Hanno Schumacher; Bernd Fischer

doi:10.1109/ISBI.2013.6556539

A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields

Jan Rühaak, Lars König, Marc Hallmann, Nils Papenberg, Stefan Heldmann, Hanno Schumacher, Bernd Fischer

Abstract

We present a super fast variational algorithm for the challenging problem of multimodal image registration. It is capable of registering full-body CT and PET images in about a second on a standard CPU with virtually no memory requirements. The algorithm is founded on a Gauss-Newton optimization scheme with specifically tailored, mathematically optimized computations for objective function and derivatives. It is fully parallelized and perfectly scalable, thus directly suitable for usage in many-core environments. The accuracy of our method was tested on 21 PET-CT scan pairs from clinical routine. The method was able to correct random distortions in the range from -10 cm to 10 cm translation and from -15° to 15° degree rotation to subvoxel accuracy. In addition, it exhibits excellent robustness to noise.

Original language	English
Title of host publication	2013 IEEE 10th International Symposium on Biomedical Imaging
Number of pages	4
Place of Publication	San Francisco, California, USA
Publisher	IEEE
Publication date	01.04.2013
Pages	572-575
ISBN (Print)	978-1-4673-6456-0
ISBN (Electronic)	978-1-4673-6455-3
DOIs	https://doi.org/10.1109/ISBI.2013.6556539
Publication status	Published - 01.04.2013
Event	2013 IEEE 10th International Symposium on Biomedical Imaging - San Francisco, United States Duration: 07.04.2014 → 11.04.2014

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ISBI.2013.6556539

http://www.mic.uni-luebeck.de/uploads/tx_wapublications/2013-ISBI-RKHPHSF.pdf

Cite this

Rühaak, J., König, L., Hallmann, M., Papenberg, N., Heldmann, S., Schumacher, H., & Fischer, B. (2013). A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields. In 2013 IEEE 10th International Symposium on Biomedical Imaging (pp. 572-575). ( International Symposium on Biomedical Imaging). IEEE. https://doi.org/10.1109/ISBI.2013.6556539

@inproceedings{5f3319810e3446f9852b56a25dd47d88,

title = "A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields",

abstract = "We present a super fast variational algorithm for the challenging problem of multimodal image registration. It is capable of registering full-body CT and PET images in about a second on a standard CPU with virtually no memory requirements. The algorithm is founded on a Gauss-Newton optimization scheme with specifically tailored, mathematically optimized computations for objective function and derivatives. It is fully parallelized and perfectly scalable, thus directly suitable for usage in many-core environments. The accuracy of our method was tested on 21 PET-CT scan pairs from clinical routine. The method was able to correct random distortions in the range from -10 cm to 10 cm translation and from -15° to 15° degree rotation to subvoxel accuracy. In addition, it exhibits excellent robustness to noise.",

author = "Jan R{\"u}haak and Lars K{\"o}nig and Marc Hallmann and Nils Papenberg and Stefan Heldmann and Hanno Schumacher and Bernd Fischer",

year = "2013",

month = apr,

day = "1",

doi = "10.1109/ISBI.2013.6556539",

language = "English",

isbn = "978-1-4673-6456-0",

series = " International Symposium on Biomedical Imaging",

publisher = "IEEE",

pages = "572--575",

booktitle = "2013 IEEE 10th International Symposium on Biomedical Imaging",

note = "2013 IEEE 10th International Symposium on Biomedical Imaging ; Conference date: 07-04-2014 Through 11-04-2014",

}

Rühaak, J, König, L, Hallmann, M, Papenberg, N , Heldmann, S, Schumacher, H & Fischer, B 2013, A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields. in 2013 IEEE 10th International Symposium on Biomedical Imaging. International Symposium on Biomedical Imaging, IEEE, San Francisco, California, USA, pp. 572-575, 2013 IEEE 10th International Symposium on Biomedical Imaging, San Francisco, California, United States, 07.04.14. https://doi.org/10.1109/ISBI.2013.6556539

A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields. / Rühaak, Jan; König, Lars; Hallmann, Marc et al.
2013 IEEE 10th International Symposium on Biomedical Imaging. San Francisco, California, USA: IEEE, 2013. p. 572-575 ( International Symposium on Biomedical Imaging).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields

AU - Rühaak, Jan

AU - König, Lars

AU - Hallmann, Marc

AU - Papenberg, Nils

AU - Heldmann, Stefan

AU - Schumacher, Hanno

AU - Fischer, Bernd

PY - 2013/4/1

Y1 - 2013/4/1

N2 - We present a super fast variational algorithm for the challenging problem of multimodal image registration. It is capable of registering full-body CT and PET images in about a second on a standard CPU with virtually no memory requirements. The algorithm is founded on a Gauss-Newton optimization scheme with specifically tailored, mathematically optimized computations for objective function and derivatives. It is fully parallelized and perfectly scalable, thus directly suitable for usage in many-core environments. The accuracy of our method was tested on 21 PET-CT scan pairs from clinical routine. The method was able to correct random distortions in the range from -10 cm to 10 cm translation and from -15° to 15° degree rotation to subvoxel accuracy. In addition, it exhibits excellent robustness to noise.

AB - We present a super fast variational algorithm for the challenging problem of multimodal image registration. It is capable of registering full-body CT and PET images in about a second on a standard CPU with virtually no memory requirements. The algorithm is founded on a Gauss-Newton optimization scheme with specifically tailored, mathematically optimized computations for objective function and derivatives. It is fully parallelized and perfectly scalable, thus directly suitable for usage in many-core environments. The accuracy of our method was tested on 21 PET-CT scan pairs from clinical routine. The method was able to correct random distortions in the range from -10 cm to 10 cm translation and from -15° to 15° degree rotation to subvoxel accuracy. In addition, it exhibits excellent robustness to noise.

U2 - 10.1109/ISBI.2013.6556539

DO - 10.1109/ISBI.2013.6556539

M3 - Conference contribution

SN - 978-1-4673-6456-0

T3 - International Symposium on Biomedical Imaging

SP - 572

EP - 575

BT - 2013 IEEE 10th International Symposium on Biomedical Imaging

PB - IEEE

CY - San Francisco, California, USA

T2 - 2013 IEEE 10th International Symposium on Biomedical Imaging

Y2 - 7 April 2014 through 11 April 2014

ER -

Rühaak J, König L, Hallmann M, Papenberg N , Heldmann S, Schumacher H et al. A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields. In 2013 IEEE 10th International Symposium on Biomedical Imaging. San Francisco, California, USA: IEEE. 2013. p. 572-575. ( International Symposium on Biomedical Imaging). doi: 10.1109/ISBI.2013.6556539

A Fully Parallel Algorithm for Multimodal Image Registration Using Normalized Gradient Fields

Abstract

UN SDGs

Access to Document

Fingerprint

Cite this