Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images

Concetta Piazzese; M. Chiara Carminati; Andrea Colombo; Rolf Krause; Mark Potse; Lynn Weinert; Gloria Tamborini; Mauro Pepi; Roberto M. Lang; Enrico G. Caiani

doi:10.1109/CIC.2015.7408597

Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images

Concetta Piazzese, M. Chiara Carminati, Andrea Colombo, Rolf Krause, Mark Potse, Lynn Weinert, Gloria Tamborini, Mauro Pepi, Roberto M. Lang, Enrico G. Caiani

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

Statistical shape models (SSMs) represent a powerful tool used in patient-specific modeling to segment medical images because they incorporate a-priori knowledge that guide the model during deformation. Our aim was to evaluate segmentation accuracy in terms of left ventricular (LV) volumes obtained using four different SSMs versus manual gold standard tracing on cardiac magnetic resonance (CMR) images. A database of 3D echocardiographic (3DE) LV surfaces obtained in 435 patients was used to generate four different SSMs, based on cardiac phase selection. Each model was scaled and deformed to detect LV endocardial contours in the end-diastolic (ED) and end-systolic (ES) frames of a CMR short-axis (SAX) stack for 15 patients with normal LV function. Linear correlation and Bland-Altman analyses versus gold-standard showed in all cases high correlation (r2>0.95), non-significant biases and narrow limits of agreement.

Original language	English
Title of host publication	Computing in Cardiology Conference 2015, CinC 2015
Editors	Alan Murray
Publisher	IEEE Computer Society
Pages	105-108
Number of pages	4
Volume	42
ISBN (Electronic)	9781509006847
ISBN (Print)	9781509006854
DOIs	https://doi.org/10.1109/CIC.2015.7408597
Publication status	Published - 6 Sep 2015
Externally published	Yes
Event	42nd Computing in Cardiology Conference - Nice, France Duration: 6 Sep 2015 → 9 Sep 2015 Conference number: 42 https://www.cinc.org/past-computing-in-cardiology-conferences/ http://www.i3s.unice.fr/cinc2015/

Publication series

Name	Computing in Cardiology
Volume	42
ISSN (Print)	2325-8861
ISSN (Electronic)	2325-887X

Conference

Conference	42nd Computing in Cardiology Conference
Abbreviated title	CinC 2015
Country/Territory	France
City	Nice
Period	6/09/15 → 9/09/15
Internet address	https://www.cinc.org/past-computing-in-cardiology-conferences/ http://www.i3s.unice.fr/cinc2015/

Access to Document

10.1109/CIC.2015.7408597Licence: CC BY

Cite this

Piazzese, C., Carminati, M. C., Colombo, A., Krause, R., Potse, M., Weinert, L., Tamborini, G., Pepi, M., Lang, R. M., & Caiani, E. G. (2015). Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images. In A. Murray (Ed.), Computing in Cardiology Conference 2015, CinC 2015 (Vol. 42, pp. 105-108). [7408597] (Computing in Cardiology; Vol. 42). IEEE Computer Society. https://doi.org/10.1109/CIC.2015.7408597

@inproceedings{a3eb02c816cd417f87461dea2bf6c3e1,

title = "Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images",

abstract = "Statistical shape models (SSMs) represent a powerful tool used in patient-specific modeling to segment medical images because they incorporate a-priori knowledge that guide the model during deformation. Our aim was to evaluate segmentation accuracy in terms of left ventricular (LV) volumes obtained using four different SSMs versus manual gold standard tracing on cardiac magnetic resonance (CMR) images. A database of 3D echocardiographic (3DE) LV surfaces obtained in 435 patients was used to generate four different SSMs, based on cardiac phase selection. Each model was scaled and deformed to detect LV endocardial contours in the end-diastolic (ED) and end-systolic (ES) frames of a CMR short-axis (SAX) stack for 15 patients with normal LV function. Linear correlation and Bland-Altman analyses versus gold-standard showed in all cases high correlation (r2>0.95), non-significant biases and narrow limits of agreement.",

keywords = "Image segmentation, Three-dimensional displays, Biomedical imaging, Magnetic resonance imaging, Motion segmentation, Image edge detection, Statistical analysis",

author = "Concetta Piazzese and Carminati, {M. Chiara} and Andrea Colombo and Rolf Krause and Mark Potse and Lynn Weinert and Gloria Tamborini and Mauro Pepi and Lang, {Roberto M.} and Caiani, {Enrico G.}",

year = "2015",

month = sep,

day = "6",

doi = "10.1109/CIC.2015.7408597",

language = "English",

isbn = "9781509006854",

volume = "42",

series = "Computing in Cardiology",

publisher = "IEEE Computer Society",

pages = "105--108",

editor = "Alan Murray",

booktitle = "Computing in Cardiology Conference 2015, CinC 2015",

address = "United States",

note = "42nd Computing in Cardiology Conference, CinC 2015 ; Conference date: 06-09-2015 Through 09-09-2015",

url = "https://www.cinc.org/past-computing-in-cardiology-conferences/, http://www.i3s.unice.fr/cinc2015/",

}

Piazzese, C, Carminati, MC, Colombo, A, Krause, R, Potse, M, Weinert, L, Tamborini, G, Pepi, M, Lang, RM & Caiani, EG 2015, Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images. in A Murray (ed.), Computing in Cardiology Conference 2015, CinC 2015. vol. 42, 7408597, Computing in Cardiology, vol. 42, IEEE Computer Society, pp. 105-108, 42nd Computing in Cardiology Conference, Nice, France, 6/09/15. https://doi.org/10.1109/CIC.2015.7408597

Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images. / Piazzese, Concetta; Carminati, M. Chiara; Colombo, Andrea et al.

Computing in Cardiology Conference 2015, CinC 2015. ed. / Alan Murray. Vol. 42 IEEE Computer Society, 2015. p. 105-108 7408597 (Computing in Cardiology; Vol. 42).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images

AU - Piazzese, Concetta

AU - Carminati, M. Chiara

AU - Colombo, Andrea

AU - Krause, Rolf

AU - Potse, Mark

AU - Weinert, Lynn

AU - Tamborini, Gloria

AU - Pepi, Mauro

AU - Lang, Roberto M.

AU - Caiani, Enrico G.

N1 - Conference code: 42

PY - 2015/9/6

Y1 - 2015/9/6

N2 - Statistical shape models (SSMs) represent a powerful tool used in patient-specific modeling to segment medical images because they incorporate a-priori knowledge that guide the model during deformation. Our aim was to evaluate segmentation accuracy in terms of left ventricular (LV) volumes obtained using four different SSMs versus manual gold standard tracing on cardiac magnetic resonance (CMR) images. A database of 3D echocardiographic (3DE) LV surfaces obtained in 435 patients was used to generate four different SSMs, based on cardiac phase selection. Each model was scaled and deformed to detect LV endocardial contours in the end-diastolic (ED) and end-systolic (ES) frames of a CMR short-axis (SAX) stack for 15 patients with normal LV function. Linear correlation and Bland-Altman analyses versus gold-standard showed in all cases high correlation (r2>0.95), non-significant biases and narrow limits of agreement.

AB - Statistical shape models (SSMs) represent a powerful tool used in patient-specific modeling to segment medical images because they incorporate a-priori knowledge that guide the model during deformation. Our aim was to evaluate segmentation accuracy in terms of left ventricular (LV) volumes obtained using four different SSMs versus manual gold standard tracing on cardiac magnetic resonance (CMR) images. A database of 3D echocardiographic (3DE) LV surfaces obtained in 435 patients was used to generate four different SSMs, based on cardiac phase selection. Each model was scaled and deformed to detect LV endocardial contours in the end-diastolic (ED) and end-systolic (ES) frames of a CMR short-axis (SAX) stack for 15 patients with normal LV function. Linear correlation and Bland-Altman analyses versus gold-standard showed in all cases high correlation (r2>0.95), non-significant biases and narrow limits of agreement.

KW - Image segmentation

KW - Three-dimensional displays

KW - Biomedical imaging

KW - Magnetic resonance imaging

KW - Motion segmentation

KW - Image edge detection

KW - Statistical analysis

UR - http://www.scopus.com/inward/record.url?scp=84964070462&partnerID=8YFLogxK

U2 - 10.1109/CIC.2015.7408597

DO - 10.1109/CIC.2015.7408597

M3 - Conference contribution

AN - SCOPUS:84964070462

SN - 9781509006854

VL - 42

T3 - Computing in Cardiology

SP - 105

EP - 108

BT - Computing in Cardiology Conference 2015, CinC 2015

A2 - Murray, Alan

PB - IEEE Computer Society

T2 - 42nd Computing in Cardiology Conference

Y2 - 6 September 2015 through 9 September 2015

ER -

Piazzese C, Carminati MC, Colombo A, Krause R, Potse M, Weinert L et al. Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images. In Murray A, editor, Computing in Cardiology Conference 2015, CinC 2015. Vol. 42. IEEE Computer Society. 2015. p. 105-108. 7408597. (Computing in Cardiology). doi: 10.1109/CIC.2015.7408597

Evaluation of different statistical shape models for segmentation of the left ventricular endocardium from magnetic resonance images

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint