Selecting registration schemes in case of interstitial lung disease follow-up in CT

Georgios Vlachopoulos, Panagiotis Korfiatis, Spyros Skiadopoulos, Alexandra Kazantzi, Christina Kalogeropoulou, Ioannis Pratikakis, Lena Costaridou

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Purpose: Primary goal of this study is to select optimal registration schemes in the framework of interstitial lung disease (ILD) follow-up analysis in CT. Methods: A set of 128 multiresolution schemes composed of multiresolution nonrigid and combinations of rigid and nonrigid registration schemes are evaluated, utilizing ten artificially warped ILD follow-up volumes, originating from ten clinical volumetric CT scans of ILD affected patients,to select candidate optimal schemes. Specifically, all combinations of four transformation models (three rigid: rigid, similarity, affine and one nonrigid: third order B-spline), four cost functions (sum-of-square distances, normalized correlation coefficient, mutual information, and normalized mutual information), four gradient descent optimizers (standard, regular step, adaptive stochastic, and finite difference), and two types of pyramids (recursive and Gaussian-smoothing) were considered. The selection process involves two stages. The first stage involves identification of schemes with deformation field singularities, according to the determinant of the Jacobian matrix. In the second stage, evaluation methodology is based on distance between corresponding landmark points in both normal lung parenchyma (NLP) and ILD affected regions. Statistical analysis was performed in order to select near optimal registration schemes per evaluation metric. Performance of the candidate registration schemes was verified on a case sample of ten clinical follow-up CT scans to obtain the selected registration schemes. Results: By considering near optimal schemes common to all ranking lists, 16 out of 128 registration schemes were initially selected. These schemes obtained submillimeter registration accuracies in terms of average distance errors 0.18-0.01 mm for NLP and 0.20-0.01 mm for ILD, in case of artificially generated follow-up data. Registration accuracy in terms of average distance error in clinical follow-up data was in the range of 1.985-2.156 mm and 1.966-2.234 mm, for NLP and ILD affected regions, respectively, excluding schemes with statistically significant lower performance (Wilcoxon signed-ranks test, p ≤ 0.05), resulting in 13 finally selected registration schemes. Conclusions: Selected registration schemes in case of ILD CT follow-up analysis indicate the significance of adaptive stochastic gradient descent optimizer, as well as the importance of combined rigid and nonrigid schemes providing high accuracy and time efficiency. The selected optimal deformable registration schemes are equivalent in terms of their accuracy and thus compatible in terms of their clinical outcome.

Original languageEnglish (US)
Pages (from-to)4511-4525
Number of pages15
JournalMedical physics
Volume42
Issue number8
DOIs
StatePublished - Aug 1 2015
Externally publishedYes

Fingerprint

Interstitial Lung Diseases
Lung
Cone-Beam Computed Tomography
Nonparametric Statistics
Costs and Cost Analysis

Keywords

  • artificially warped data
  • clinical follow-up CT data
  • evaluation of registration accuracy
  • interstitial lung disease
  • lung CT registration
  • quantitative image registration metrics
  • rigid and nonrigid image registration

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Vlachopoulos, G., Korfiatis, P., Skiadopoulos, S., Kazantzi, A., Kalogeropoulou, C., Pratikakis, I., & Costaridou, L. (2015). Selecting registration schemes in case of interstitial lung disease follow-up in CT. Medical physics, 42(8), 4511-4525. https://doi.org/10.1118/1.4923170

Selecting registration schemes in case of interstitial lung disease follow-up in CT. / Vlachopoulos, Georgios; Korfiatis, Panagiotis; Skiadopoulos, Spyros; Kazantzi, Alexandra; Kalogeropoulou, Christina; Pratikakis, Ioannis; Costaridou, Lena.

In: Medical physics, Vol. 42, No. 8, 01.08.2015, p. 4511-4525.

Research output: Contribution to journalArticle

Vlachopoulos, G, Korfiatis, P, Skiadopoulos, S, Kazantzi, A, Kalogeropoulou, C, Pratikakis, I & Costaridou, L 2015, 'Selecting registration schemes in case of interstitial lung disease follow-up in CT', Medical physics, vol. 42, no. 8, pp. 4511-4525. https://doi.org/10.1118/1.4923170
Vlachopoulos G, Korfiatis P, Skiadopoulos S, Kazantzi A, Kalogeropoulou C, Pratikakis I et al. Selecting registration schemes in case of interstitial lung disease follow-up in CT. Medical physics. 2015 Aug 1;42(8):4511-4525. https://doi.org/10.1118/1.4923170
Vlachopoulos, Georgios ; Korfiatis, Panagiotis ; Skiadopoulos, Spyros ; Kazantzi, Alexandra ; Kalogeropoulou, Christina ; Pratikakis, Ioannis ; Costaridou, Lena. / Selecting registration schemes in case of interstitial lung disease follow-up in CT. In: Medical physics. 2015 ; Vol. 42, No. 8. pp. 4511-4525.
@article{850c93c993584a199c26442d297fc0db,
title = "Selecting registration schemes in case of interstitial lung disease follow-up in CT",
abstract = "Purpose: Primary goal of this study is to select optimal registration schemes in the framework of interstitial lung disease (ILD) follow-up analysis in CT. Methods: A set of 128 multiresolution schemes composed of multiresolution nonrigid and combinations of rigid and nonrigid registration schemes are evaluated, utilizing ten artificially warped ILD follow-up volumes, originating from ten clinical volumetric CT scans of ILD affected patients,to select candidate optimal schemes. Specifically, all combinations of four transformation models (three rigid: rigid, similarity, affine and one nonrigid: third order B-spline), four cost functions (sum-of-square distances, normalized correlation coefficient, mutual information, and normalized mutual information), four gradient descent optimizers (standard, regular step, adaptive stochastic, and finite difference), and two types of pyramids (recursive and Gaussian-smoothing) were considered. The selection process involves two stages. The first stage involves identification of schemes with deformation field singularities, according to the determinant of the Jacobian matrix. In the second stage, evaluation methodology is based on distance between corresponding landmark points in both normal lung parenchyma (NLP) and ILD affected regions. Statistical analysis was performed in order to select near optimal registration schemes per evaluation metric. Performance of the candidate registration schemes was verified on a case sample of ten clinical follow-up CT scans to obtain the selected registration schemes. Results: By considering near optimal schemes common to all ranking lists, 16 out of 128 registration schemes were initially selected. These schemes obtained submillimeter registration accuracies in terms of average distance errors 0.18-0.01 mm for NLP and 0.20-0.01 mm for ILD, in case of artificially generated follow-up data. Registration accuracy in terms of average distance error in clinical follow-up data was in the range of 1.985-2.156 mm and 1.966-2.234 mm, for NLP and ILD affected regions, respectively, excluding schemes with statistically significant lower performance (Wilcoxon signed-ranks test, p ≤ 0.05), resulting in 13 finally selected registration schemes. Conclusions: Selected registration schemes in case of ILD CT follow-up analysis indicate the significance of adaptive stochastic gradient descent optimizer, as well as the importance of combined rigid and nonrigid schemes providing high accuracy and time efficiency. The selected optimal deformable registration schemes are equivalent in terms of their accuracy and thus compatible in terms of their clinical outcome.",
keywords = "artificially warped data, clinical follow-up CT data, evaluation of registration accuracy, interstitial lung disease, lung CT registration, quantitative image registration metrics, rigid and nonrigid image registration",
author = "Georgios Vlachopoulos and Panagiotis Korfiatis and Spyros Skiadopoulos and Alexandra Kazantzi and Christina Kalogeropoulou and Ioannis Pratikakis and Lena Costaridou",
year = "2015",
month = "8",
day = "1",
doi = "10.1118/1.4923170",
language = "English (US)",
volume = "42",
pages = "4511--4525",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "8",

}

TY - JOUR

T1 - Selecting registration schemes in case of interstitial lung disease follow-up in CT

AU - Vlachopoulos, Georgios

AU - Korfiatis, Panagiotis

AU - Skiadopoulos, Spyros

AU - Kazantzi, Alexandra

AU - Kalogeropoulou, Christina

AU - Pratikakis, Ioannis

AU - Costaridou, Lena

PY - 2015/8/1

Y1 - 2015/8/1

N2 - Purpose: Primary goal of this study is to select optimal registration schemes in the framework of interstitial lung disease (ILD) follow-up analysis in CT. Methods: A set of 128 multiresolution schemes composed of multiresolution nonrigid and combinations of rigid and nonrigid registration schemes are evaluated, utilizing ten artificially warped ILD follow-up volumes, originating from ten clinical volumetric CT scans of ILD affected patients,to select candidate optimal schemes. Specifically, all combinations of four transformation models (three rigid: rigid, similarity, affine and one nonrigid: third order B-spline), four cost functions (sum-of-square distances, normalized correlation coefficient, mutual information, and normalized mutual information), four gradient descent optimizers (standard, regular step, adaptive stochastic, and finite difference), and two types of pyramids (recursive and Gaussian-smoothing) were considered. The selection process involves two stages. The first stage involves identification of schemes with deformation field singularities, according to the determinant of the Jacobian matrix. In the second stage, evaluation methodology is based on distance between corresponding landmark points in both normal lung parenchyma (NLP) and ILD affected regions. Statistical analysis was performed in order to select near optimal registration schemes per evaluation metric. Performance of the candidate registration schemes was verified on a case sample of ten clinical follow-up CT scans to obtain the selected registration schemes. Results: By considering near optimal schemes common to all ranking lists, 16 out of 128 registration schemes were initially selected. These schemes obtained submillimeter registration accuracies in terms of average distance errors 0.18-0.01 mm for NLP and 0.20-0.01 mm for ILD, in case of artificially generated follow-up data. Registration accuracy in terms of average distance error in clinical follow-up data was in the range of 1.985-2.156 mm and 1.966-2.234 mm, for NLP and ILD affected regions, respectively, excluding schemes with statistically significant lower performance (Wilcoxon signed-ranks test, p ≤ 0.05), resulting in 13 finally selected registration schemes. Conclusions: Selected registration schemes in case of ILD CT follow-up analysis indicate the significance of adaptive stochastic gradient descent optimizer, as well as the importance of combined rigid and nonrigid schemes providing high accuracy and time efficiency. The selected optimal deformable registration schemes are equivalent in terms of their accuracy and thus compatible in terms of their clinical outcome.

AB - Purpose: Primary goal of this study is to select optimal registration schemes in the framework of interstitial lung disease (ILD) follow-up analysis in CT. Methods: A set of 128 multiresolution schemes composed of multiresolution nonrigid and combinations of rigid and nonrigid registration schemes are evaluated, utilizing ten artificially warped ILD follow-up volumes, originating from ten clinical volumetric CT scans of ILD affected patients,to select candidate optimal schemes. Specifically, all combinations of four transformation models (three rigid: rigid, similarity, affine and one nonrigid: third order B-spline), four cost functions (sum-of-square distances, normalized correlation coefficient, mutual information, and normalized mutual information), four gradient descent optimizers (standard, regular step, adaptive stochastic, and finite difference), and two types of pyramids (recursive and Gaussian-smoothing) were considered. The selection process involves two stages. The first stage involves identification of schemes with deformation field singularities, according to the determinant of the Jacobian matrix. In the second stage, evaluation methodology is based on distance between corresponding landmark points in both normal lung parenchyma (NLP) and ILD affected regions. Statistical analysis was performed in order to select near optimal registration schemes per evaluation metric. Performance of the candidate registration schemes was verified on a case sample of ten clinical follow-up CT scans to obtain the selected registration schemes. Results: By considering near optimal schemes common to all ranking lists, 16 out of 128 registration schemes were initially selected. These schemes obtained submillimeter registration accuracies in terms of average distance errors 0.18-0.01 mm for NLP and 0.20-0.01 mm for ILD, in case of artificially generated follow-up data. Registration accuracy in terms of average distance error in clinical follow-up data was in the range of 1.985-2.156 mm and 1.966-2.234 mm, for NLP and ILD affected regions, respectively, excluding schemes with statistically significant lower performance (Wilcoxon signed-ranks test, p ≤ 0.05), resulting in 13 finally selected registration schemes. Conclusions: Selected registration schemes in case of ILD CT follow-up analysis indicate the significance of adaptive stochastic gradient descent optimizer, as well as the importance of combined rigid and nonrigid schemes providing high accuracy and time efficiency. The selected optimal deformable registration schemes are equivalent in terms of their accuracy and thus compatible in terms of their clinical outcome.

KW - artificially warped data

KW - clinical follow-up CT data

KW - evaluation of registration accuracy

KW - interstitial lung disease

KW - lung CT registration

KW - quantitative image registration metrics

KW - rigid and nonrigid image registration

UR - http://www.scopus.com/inward/record.url?scp=84936791444&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936791444&partnerID=8YFLogxK

U2 - 10.1118/1.4923170

DO - 10.1118/1.4923170

M3 - Article

C2 - 26233180

AN - SCOPUS:84936791444

VL - 42

SP - 4511

EP - 4525

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 8

ER -