Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms

Masuma Akter, Andrew J. Lambert, Mark R. Pickering, Jenne M. Scarvell, Paul N. Smith, Fariha Tabassuma, Mahamud Tariq Rashid

Research output: A Conference proceeding or a Chapter in BookConference contribution

Abstract

Automatic image registration algorithms that rely on a gradient descent based approach may fail when the initial misalignment between objects is large. The registration task is even more difficult for multi-modal images because of the non-linear relationship between the pixel intensities in the images to be aligned. In this paper we will present a multi-modal image registration algorithm which successfully registers 3D CT to 2D fluoroscopy data for large initial displacements between the images. The approach uses the conditional means (CM) of the joint probability distribution of the images to establish a model linear relationship between the pixel intensities of the images and then applies log-polar transforms (LPT) in the frequency domain to estimate the in-plane scale and rotation changes between the images. Our experimental results show that the proposed approach can increase the range of initial displacements for which the algorithm is able successfully register images by a factor of 4 when compared to the best of the existing gradient-descent based approaches.

Original languageEnglish
Title of host publicationICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISBN (Electronic)9781509057856
DOIs
Publication statusPublished - 15 Mar 2017
Externally publishedYes
Event2nd International Conference on Electrical, Computer and Telecommunication Engineering, ICECTE 2016 - Rajshahi, Bangladesh
Duration: 8 Dec 201610 Dec 2016

Conference

Conference2nd International Conference on Electrical, Computer and Telecommunication Engineering, ICECTE 2016
CountryBangladesh
CityRajshahi
Period8/12/1610/12/16

Fingerprint

fluoroscopy
Image registration
Pixels
Probability distributions
registers
descent
pixels
gradients
misalignment

Cite this

Akter, M., Lambert, A. J., Pickering, M. R., Scarvell, J. M., Smith, P. N., Tabassuma, F., & Rashid, M. T. (2017). Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms. In ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering [7879600] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICECTE.2016.7879600
Akter, Masuma ; Lambert, Andrew J. ; Pickering, Mark R. ; Scarvell, Jenne M. ; Smith, Paul N. ; Tabassuma, Fariha ; Rashid, Mahamud Tariq. / Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms. ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering. IEEE, Institute of Electrical and Electronics Engineers, 2017.
@inproceedings{24b20eca65c246a7bf331c63a8af2191,
title = "Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms",
abstract = "Automatic image registration algorithms that rely on a gradient descent based approach may fail when the initial misalignment between objects is large. The registration task is even more difficult for multi-modal images because of the non-linear relationship between the pixel intensities in the images to be aligned. In this paper we will present a multi-modal image registration algorithm which successfully registers 3D CT to 2D fluoroscopy data for large initial displacements between the images. The approach uses the conditional means (CM) of the joint probability distribution of the images to establish a model linear relationship between the pixel intensities of the images and then applies log-polar transforms (LPT) in the frequency domain to estimate the in-plane scale and rotation changes between the images. Our experimental results show that the proposed approach can increase the range of initial displacements for which the algorithm is able successfully register images by a factor of 4 when compared to the best of the existing gradient-descent based approaches.",
author = "Masuma Akter and Lambert, {Andrew J.} and Pickering, {Mark R.} and Scarvell, {Jenne M.} and Smith, {Paul N.} and Fariha Tabassuma and Rashid, {Mahamud Tariq}",
year = "2017",
month = "3",
day = "15",
doi = "10.1109/ICECTE.2016.7879600",
language = "English",
booktitle = "ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Akter, M, Lambert, AJ, Pickering, MR, Scarvell, JM, Smith, PN, Tabassuma, F & Rashid, MT 2017, Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms. in ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering., 7879600, IEEE, Institute of Electrical and Electronics Engineers, 2nd International Conference on Electrical, Computer and Telecommunication Engineering, ICECTE 2016, Rajshahi, Bangladesh, 8/12/16. https://doi.org/10.1109/ICECTE.2016.7879600

Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms. / Akter, Masuma; Lambert, Andrew J.; Pickering, Mark R.; Scarvell, Jenne M.; Smith, Paul N.; Tabassuma, Fariha; Rashid, Mahamud Tariq.

ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering. IEEE, Institute of Electrical and Electronics Engineers, 2017. 7879600.

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms

AU - Akter, Masuma

AU - Lambert, Andrew J.

AU - Pickering, Mark R.

AU - Scarvell, Jenne M.

AU - Smith, Paul N.

AU - Tabassuma, Fariha

AU - Rashid, Mahamud Tariq

PY - 2017/3/15

Y1 - 2017/3/15

N2 - Automatic image registration algorithms that rely on a gradient descent based approach may fail when the initial misalignment between objects is large. The registration task is even more difficult for multi-modal images because of the non-linear relationship between the pixel intensities in the images to be aligned. In this paper we will present a multi-modal image registration algorithm which successfully registers 3D CT to 2D fluoroscopy data for large initial displacements between the images. The approach uses the conditional means (CM) of the joint probability distribution of the images to establish a model linear relationship between the pixel intensities of the images and then applies log-polar transforms (LPT) in the frequency domain to estimate the in-plane scale and rotation changes between the images. Our experimental results show that the proposed approach can increase the range of initial displacements for which the algorithm is able successfully register images by a factor of 4 when compared to the best of the existing gradient-descent based approaches.

AB - Automatic image registration algorithms that rely on a gradient descent based approach may fail when the initial misalignment between objects is large. The registration task is even more difficult for multi-modal images because of the non-linear relationship between the pixel intensities in the images to be aligned. In this paper we will present a multi-modal image registration algorithm which successfully registers 3D CT to 2D fluoroscopy data for large initial displacements between the images. The approach uses the conditional means (CM) of the joint probability distribution of the images to establish a model linear relationship between the pixel intensities of the images and then applies log-polar transforms (LPT) in the frequency domain to estimate the in-plane scale and rotation changes between the images. Our experimental results show that the proposed approach can increase the range of initial displacements for which the algorithm is able successfully register images by a factor of 4 when compared to the best of the existing gradient-descent based approaches.

UR - http://www.scopus.com/inward/record.url?scp=85017310635&partnerID=8YFLogxK

U2 - 10.1109/ICECTE.2016.7879600

DO - 10.1109/ICECTE.2016.7879600

M3 - Conference contribution

BT - ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering

PB - IEEE, Institute of Electrical and Electronics Engineers

ER -

Akter M, Lambert AJ, Pickering MR, Scarvell JM, Smith PN, Tabassuma F et al. Improved robustness of 3D CT to 2D fluoroscopy image registration using log polar transforms. In ICECTE 2016 - 2nd International Conference on Electrical, Computer and Telecommunication Engineering. IEEE, Institute of Electrical and Electronics Engineers. 2017. 7879600 https://doi.org/10.1109/ICECTE.2016.7879600