3D right ventricular endocardium segmentation in cardiac magnetic resonance images by using a new inter-modality statistical shape modelling method

Concetta Piazzese, M. Chiara Carminati, Rolf Krause, Angelo Auricchio, Lynn Weinert, Paola Gripari, Gloria Tamborini, Gianluca Pontone, Daniele Andreini, Roberto M. Lang, Mauro Pepi, Enrico G. Caiani

Research output: Contribution to journalArticle

Abstract

Objective: Statistical shape modelling (SSM) has established as a powerful method for segmenting the left ventricle in cardiac magnetic resonance (CMR) images However, applying them to segment the right ventricle (RV) is not straightforward because of the complex structure of this chamber. Our aim was to develop a new inter-modality SSM-based approach to detect the RV endocardium in CMR data. Methods: Real-time transthoracic 3D echocardiographic (3DE) images of 219 retrospective patients were used to populate a large database containing 4347 3D RV surfaces and train a model. The initial position, orientation and scale of the model in the CMR stack were semi-automatically derived. The detection process consisted in iteratively deforming the model to match endocardial borders in each CMR plane until convergence was reached. Clinical values obtained with the presented SSM method were compared with gold-standard (GS) corresponding parameters. Results: CMR images of 50 patients with different pathologies were used to test the proposed segmentation method. Average processing time was 2 min (including manual initialization) per patient. High correlations (r2 > 0.76) and not significant bias (Bland-Altman analysis) were observed when evaluating clinical parameters. Quantitative analysis showed high values of Dice coefficient (0.87 ± 0.03), acceptable Hausdorff distance (9.35 ± 1.51 mm) and small point-to-surface distance (1.91 ± 0.26 mm). Conclusion: A novel SSM-based approach to segment the RV endocardium in CMR scans by using a model trained on 3DE-derived RV endocardial surfaces, was proposed. This inter-modality technique proved to be rapid when segmenting the RV endocardium with an accurate anatomical delineation, in particular in apical and basal regions.

Original languageEnglish
Article number101866
JournalBiomedical Signal Processing and Control
Volume58
DOIs
Publication statusPublished - Apr 2020

Fingerprint

Endocardium
Magnetic resonance
Heart Ventricles
Magnetic Resonance Spectroscopy
Pathology
Gold
Databases
Processing
Chemical analysis

Keywords

  • Cardiac magnetic resonance images
  • Image segmentation
  • Right ventricular volume
  • Statistical shape model

ASJC Scopus subject areas

  • Signal Processing
  • Health Informatics

Cite this

3D right ventricular endocardium segmentation in cardiac magnetic resonance images by using a new inter-modality statistical shape modelling method. / Piazzese, Concetta; Carminati, M. Chiara; Krause, Rolf; Auricchio, Angelo; Weinert, Lynn; Gripari, Paola; Tamborini, Gloria; Pontone, Gianluca; Andreini, Daniele; Lang, Roberto M.; Pepi, Mauro; Caiani, Enrico G.

In: Biomedical Signal Processing and Control, Vol. 58, 101866, 04.2020.

Research output: Contribution to journalArticle

@article{27285126ceee4c1da88b3328c69a067a,
title = "3D right ventricular endocardium segmentation in cardiac magnetic resonance images by using a new inter-modality statistical shape modelling method",
abstract = "Objective: Statistical shape modelling (SSM) has established as a powerful method for segmenting the left ventricle in cardiac magnetic resonance (CMR) images However, applying them to segment the right ventricle (RV) is not straightforward because of the complex structure of this chamber. Our aim was to develop a new inter-modality SSM-based approach to detect the RV endocardium in CMR data. Methods: Real-time transthoracic 3D echocardiographic (3DE) images of 219 retrospective patients were used to populate a large database containing 4347 3D RV surfaces and train a model. The initial position, orientation and scale of the model in the CMR stack were semi-automatically derived. The detection process consisted in iteratively deforming the model to match endocardial borders in each CMR plane until convergence was reached. Clinical values obtained with the presented SSM method were compared with gold-standard (GS) corresponding parameters. Results: CMR images of 50 patients with different pathologies were used to test the proposed segmentation method. Average processing time was 2 min (including manual initialization) per patient. High correlations (r2 > 0.76) and not significant bias (Bland-Altman analysis) were observed when evaluating clinical parameters. Quantitative analysis showed high values of Dice coefficient (0.87 ± 0.03), acceptable Hausdorff distance (9.35 ± 1.51 mm) and small point-to-surface distance (1.91 ± 0.26 mm). Conclusion: A novel SSM-based approach to segment the RV endocardium in CMR scans by using a model trained on 3DE-derived RV endocardial surfaces, was proposed. This inter-modality technique proved to be rapid when segmenting the RV endocardium with an accurate anatomical delineation, in particular in apical and basal regions.",
keywords = "Cardiac magnetic resonance images, Image segmentation, Right ventricular volume, Statistical shape model",
author = "Concetta Piazzese and Carminati, {M. Chiara} and Rolf Krause and Angelo Auricchio and Lynn Weinert and Paola Gripari and Gloria Tamborini and Gianluca Pontone and Daniele Andreini and Lang, {Roberto M.} and Mauro Pepi and Caiani, {Enrico G.}",
year = "2020",
month = "4",
doi = "10.1016/j.bspc.2020.101866",
language = "English",
volume = "58",
journal = "Biomedical Signal Processing and Control",
issn = "1746-8094",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - 3D right ventricular endocardium segmentation in cardiac magnetic resonance images by using a new inter-modality statistical shape modelling method

AU - Piazzese, Concetta

AU - Carminati, M. Chiara

AU - Krause, Rolf

AU - Auricchio, Angelo

AU - Weinert, Lynn

AU - Gripari, Paola

AU - Tamborini, Gloria

AU - Pontone, Gianluca

AU - Andreini, Daniele

AU - Lang, Roberto M.

AU - Pepi, Mauro

AU - Caiani, Enrico G.

PY - 2020/4

Y1 - 2020/4

N2 - Objective: Statistical shape modelling (SSM) has established as a powerful method for segmenting the left ventricle in cardiac magnetic resonance (CMR) images However, applying them to segment the right ventricle (RV) is not straightforward because of the complex structure of this chamber. Our aim was to develop a new inter-modality SSM-based approach to detect the RV endocardium in CMR data. Methods: Real-time transthoracic 3D echocardiographic (3DE) images of 219 retrospective patients were used to populate a large database containing 4347 3D RV surfaces and train a model. The initial position, orientation and scale of the model in the CMR stack were semi-automatically derived. The detection process consisted in iteratively deforming the model to match endocardial borders in each CMR plane until convergence was reached. Clinical values obtained with the presented SSM method were compared with gold-standard (GS) corresponding parameters. Results: CMR images of 50 patients with different pathologies were used to test the proposed segmentation method. Average processing time was 2 min (including manual initialization) per patient. High correlations (r2 > 0.76) and not significant bias (Bland-Altman analysis) were observed when evaluating clinical parameters. Quantitative analysis showed high values of Dice coefficient (0.87 ± 0.03), acceptable Hausdorff distance (9.35 ± 1.51 mm) and small point-to-surface distance (1.91 ± 0.26 mm). Conclusion: A novel SSM-based approach to segment the RV endocardium in CMR scans by using a model trained on 3DE-derived RV endocardial surfaces, was proposed. This inter-modality technique proved to be rapid when segmenting the RV endocardium with an accurate anatomical delineation, in particular in apical and basal regions.

AB - Objective: Statistical shape modelling (SSM) has established as a powerful method for segmenting the left ventricle in cardiac magnetic resonance (CMR) images However, applying them to segment the right ventricle (RV) is not straightforward because of the complex structure of this chamber. Our aim was to develop a new inter-modality SSM-based approach to detect the RV endocardium in CMR data. Methods: Real-time transthoracic 3D echocardiographic (3DE) images of 219 retrospective patients were used to populate a large database containing 4347 3D RV surfaces and train a model. The initial position, orientation and scale of the model in the CMR stack were semi-automatically derived. The detection process consisted in iteratively deforming the model to match endocardial borders in each CMR plane until convergence was reached. Clinical values obtained with the presented SSM method were compared with gold-standard (GS) corresponding parameters. Results: CMR images of 50 patients with different pathologies were used to test the proposed segmentation method. Average processing time was 2 min (including manual initialization) per patient. High correlations (r2 > 0.76) and not significant bias (Bland-Altman analysis) were observed when evaluating clinical parameters. Quantitative analysis showed high values of Dice coefficient (0.87 ± 0.03), acceptable Hausdorff distance (9.35 ± 1.51 mm) and small point-to-surface distance (1.91 ± 0.26 mm). Conclusion: A novel SSM-based approach to segment the RV endocardium in CMR scans by using a model trained on 3DE-derived RV endocardial surfaces, was proposed. This inter-modality technique proved to be rapid when segmenting the RV endocardium with an accurate anatomical delineation, in particular in apical and basal regions.

KW - Cardiac magnetic resonance images

KW - Image segmentation

KW - Right ventricular volume

KW - Statistical shape model

UR - http://www.scopus.com/inward/record.url?scp=85078134302&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85078134302&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2020.101866

DO - 10.1016/j.bspc.2020.101866

M3 - Article

AN - SCOPUS:85078134302

VL - 58

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

SN - 1746-8094

M1 - 101866

ER -