Ultrasound elastography using three images

Hassan Rivaz, Emad M. Boctor, Michael A. Choti, Gregory D. Hager

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Displacement1 estimation is an essential step for ultrasound elastography and numerous techniques have been proposed to improve its quality using two frames of ultrasound RF data. This paper introduces a technique for calculating a displacement field from three frames of ultrasound RF data. To this end, we first introduce constraints on variations of the displacement field with time using mechanics of materials. These constraints are then used to generate a regularized cost function that incorporates amplitude similarity of three ultrasound images and displacement continuity. We optimize the cost function in an expectation maximization (EM) framework. Iteratively reweighted least squares (IRLS) is used to minimize the effect of outliers. We show that, compared to using two images, the new algorithm reduces the noise of the displacement estimation. The displacement field is used to generate strain images for quasi-static elastography. Phantom experiments and in-vivo patient trials of imaging liver tumors and monitoring thermal ablation therapy of liver cancer are presented for validation.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages371-378
Number of pages8
Volume6891 LNCS
EditionPART 1
DOIs
StatePublished - 2011
Event14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011 - Toronto, ON, Canada
Duration: Sep 18 2011Sep 22 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume6891 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011
CountryCanada
CityToronto, ON
Period9/18/119/22/11

Fingerprint

Ultrasound
Ultrasonics
Cost functions
Liver
Cost Function
Iteratively Reweighted Least Squares
Ablation
Ultrasound Image
Expectation Maximization
Tumors
Phantom
Mechanics
Outlier
Therapy
Tumor
Cancer
Imaging techniques
Optimise
Imaging
Monitoring

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Rivaz, H., Boctor, E. M., Choti, M. A., & Hager, G. D. (2011). Ultrasound elastography using three images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 1 ed., Vol. 6891 LNCS, pp. 371-378). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6891 LNCS, No. PART 1). https://doi.org/10.1007/978-3-642-23623-5_47

Ultrasound elastography using three images. / Rivaz, Hassan; Boctor, Emad M.; Choti, Michael A.; Hager, Gregory D.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6891 LNCS PART 1. ed. 2011. p. 371-378 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6891 LNCS, No. PART 1).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rivaz, H, Boctor, EM, Choti, MA & Hager, GD 2011, Ultrasound elastography using three images. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 edn, vol. 6891 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 6891 LNCS, pp. 371-378, 14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011, Toronto, ON, Canada, 9/18/11. https://doi.org/10.1007/978-3-642-23623-5_47
Rivaz H, Boctor EM, Choti MA, Hager GD. Ultrasound elastography using three images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 ed. Vol. 6891 LNCS. 2011. p. 371-378. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-642-23623-5_47
Rivaz, Hassan ; Boctor, Emad M. ; Choti, Michael A. ; Hager, Gregory D. / Ultrasound elastography using three images. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6891 LNCS PART 1. ed. 2011. pp. 371-378 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).
@inproceedings{e7d20bfccb4147cea4c561a6a0544861,
title = "Ultrasound elastography using three images",
abstract = "Displacement1 estimation is an essential step for ultrasound elastography and numerous techniques have been proposed to improve its quality using two frames of ultrasound RF data. This paper introduces a technique for calculating a displacement field from three frames of ultrasound RF data. To this end, we first introduce constraints on variations of the displacement field with time using mechanics of materials. These constraints are then used to generate a regularized cost function that incorporates amplitude similarity of three ultrasound images and displacement continuity. We optimize the cost function in an expectation maximization (EM) framework. Iteratively reweighted least squares (IRLS) is used to minimize the effect of outliers. We show that, compared to using two images, the new algorithm reduces the noise of the displacement estimation. The displacement field is used to generate strain images for quasi-static elastography. Phantom experiments and in-vivo patient trials of imaging liver tumors and monitoring thermal ablation therapy of liver cancer are presented for validation.",
author = "Hassan Rivaz and Boctor, {Emad M.} and Choti, {Michael A.} and Hager, {Gregory D.}",
year = "2011",
doi = "10.1007/978-3-642-23623-5_47",
language = "English (US)",
isbn = "9783642236228",
volume = "6891 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 1",
pages = "371--378",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 1",

}

TY - GEN

T1 - Ultrasound elastography using three images

AU - Rivaz, Hassan

AU - Boctor, Emad M.

AU - Choti, Michael A.

AU - Hager, Gregory D.

PY - 2011

Y1 - 2011

N2 - Displacement1 estimation is an essential step for ultrasound elastography and numerous techniques have been proposed to improve its quality using two frames of ultrasound RF data. This paper introduces a technique for calculating a displacement field from three frames of ultrasound RF data. To this end, we first introduce constraints on variations of the displacement field with time using mechanics of materials. These constraints are then used to generate a regularized cost function that incorporates amplitude similarity of three ultrasound images and displacement continuity. We optimize the cost function in an expectation maximization (EM) framework. Iteratively reweighted least squares (IRLS) is used to minimize the effect of outliers. We show that, compared to using two images, the new algorithm reduces the noise of the displacement estimation. The displacement field is used to generate strain images for quasi-static elastography. Phantom experiments and in-vivo patient trials of imaging liver tumors and monitoring thermal ablation therapy of liver cancer are presented for validation.

AB - Displacement1 estimation is an essential step for ultrasound elastography and numerous techniques have been proposed to improve its quality using two frames of ultrasound RF data. This paper introduces a technique for calculating a displacement field from three frames of ultrasound RF data. To this end, we first introduce constraints on variations of the displacement field with time using mechanics of materials. These constraints are then used to generate a regularized cost function that incorporates amplitude similarity of three ultrasound images and displacement continuity. We optimize the cost function in an expectation maximization (EM) framework. Iteratively reweighted least squares (IRLS) is used to minimize the effect of outliers. We show that, compared to using two images, the new algorithm reduces the noise of the displacement estimation. The displacement field is used to generate strain images for quasi-static elastography. Phantom experiments and in-vivo patient trials of imaging liver tumors and monitoring thermal ablation therapy of liver cancer are presented for validation.

UR - http://www.scopus.com/inward/record.url?scp=80053473876&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053473876&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-23623-5_47

DO - 10.1007/978-3-642-23623-5_47

M3 - Conference contribution

C2 - 22003639

AN - SCOPUS:80053473876

SN - 9783642236228

VL - 6891 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 371

EP - 378

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -