Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation

Rahil Shahzad; Martin Fasshauer; Boudewijn P.F. Lelieveldt; Joachim Lotz; Rob J. van der Geest

doi:10.1007/978-3-319-92258-4_6

Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation

Rahil Shahzad^*, Martin Fasshauer, Boudewijn P.F. Lelieveldt, Joachim Lotz, Rob J. van der Geest

^*Corresponding author for this work

Pattern Recognition and Bioinformatics

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

Abstract

We present a workflow for processing real-time cardiac MR (RT-CMR) scans for segmenting the left ventricle (LV) on short-axis slices (SAX). Our method is based on image registration, where the LV endocardium and epicardium are segmented by propagating a reference contour over all the frames of the RT-CMR SAX scans. Our method was evaluated on 19 subjects, the accuracy of the automatic LV endocardium and epicardium segmentation was compared to those defined manually. The proposed method obtained a dice similarity coefficient (DSC) of 0.94 and a mean surface-to-surface distance (MSD) measure of 0.89 ± 0.53 mm. Additionally, a number of automatically obtained clinical measures were compared to ground truth values. On average we obtained a Pearson’s correlation coefficient (R) of 0.94 (0.99–0.74).

Original language	English
Title of host publication	Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings
Editors	Stefan Klein, Marius Staring, Stanley Durrleman, Stefan Sommer
Place of Publication	Cham
Publisher	Springer
Pages	57-66
Number of pages	10
ISBN (Electronic)	978-3-319-92258-4
ISBN (Print)	978-3-319-92257-7
DOIs	https://doi.org/10.1007/978-3-319-92258-4_6
Publication status	Published - 2018
Event	WBIR 2018: 8th International Workshop on Biomedical Image Registration - Leiden, Netherlands Duration: 28 Jun 2018 → 29 Jun 2018 Conference number: 8 https://wbir2018.nl/

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10883 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	WBIR 2018
Country/Territory	Netherlands
City	Leiden
Period	28/06/18 → 29/06/18
Internet address	https://wbir2018.nl/

Keywords

Left ventricle
Realtime-MR
Registration
Segmentation
Semi-automatic

Access to Document

10.1007/978-3-319-92258-4_6

Cite this

Shahzad, R., Fasshauer, M., Lelieveldt, B. P. F., Lotz, J., & van der Geest, R. J. (2018). Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation. In S. Klein, M. Staring, S. Durrleman, & S. Sommer (Eds.), Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings (pp. 57-66). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10883 LNCS). Springer. https://doi.org/10.1007/978-3-319-92258-4_6

Shahzad, Rahil ; Fasshauer, Martin ; Lelieveldt, Boudewijn P.F. et al. / Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation. Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings. editor / Stefan Klein ; Marius Staring ; Stanley Durrleman ; Stefan Sommer. Cham : Springer, 2018. pp. 57-66 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{394cbd42ed584e7e829c9cfb36ff0004,

title = "Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation",

abstract = "We present a workflow for processing real-time cardiac MR (RT-CMR) scans for segmenting the left ventricle (LV) on short-axis slices (SAX). Our method is based on image registration, where the LV endocardium and epicardium are segmented by propagating a reference contour over all the frames of the RT-CMR SAX scans. Our method was evaluated on 19 subjects, the accuracy of the automatic LV endocardium and epicardium segmentation was compared to those defined manually. The proposed method obtained a dice similarity coefficient (DSC) of 0.94 and a mean surface-to-surface distance (MSD) measure of 0.89 ± 0.53 mm. Additionally, a number of automatically obtained clinical measures were compared to ground truth values. On average we obtained a Pearson{\textquoteright}s correlation coefficient (R) of 0.94 (0.99–0.74).",

keywords = "Left ventricle, Realtime-MR, Registration, Segmentation, Semi-automatic",

author = "Rahil Shahzad and Martin Fasshauer and Lelieveldt, {Boudewijn P.F.} and Joachim Lotz and {van der Geest}, {Rob J.}",

year = "2018",

doi = "10.1007/978-3-319-92258-4_6",

language = "English",

isbn = "978-3-319-92257-7",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "57--66",

editor = "Stefan Klein and Marius Staring and Stanley Durrleman and Stefan Sommer",

booktitle = "Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings",

note = "WBIR 2018 : 8th International Workshop on Biomedical Image Registration ; Conference date: 28-06-2018 Through 29-06-2018",

url = "https://wbir2018.nl/",

}

Shahzad, R, Fasshauer, M, Lelieveldt, BPF, Lotz, J & van der Geest, RJ 2018, Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation. in S Klein, M Staring, S Durrleman & S Sommer (eds), Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10883 LNCS, Springer, Cham, pp. 57-66, WBIR 2018, Leiden, Netherlands, 28/06/18. https://doi.org/10.1007/978-3-319-92258-4_6

Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation. / Shahzad, Rahil; Fasshauer, Martin; Lelieveldt, Boudewijn P.F. et al.
Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings. ed. / Stefan Klein; Marius Staring; Stanley Durrleman; Stefan Sommer. Cham: Springer, 2018. p. 57-66 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10883 LNCS).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation

AU - Shahzad, Rahil

AU - Fasshauer, Martin

AU - Lelieveldt, Boudewijn P.F.

AU - Lotz, Joachim

AU - van der Geest, Rob J.

N1 - Conference code: 8

PY - 2018

Y1 - 2018

N2 - We present a workflow for processing real-time cardiac MR (RT-CMR) scans for segmenting the left ventricle (LV) on short-axis slices (SAX). Our method is based on image registration, where the LV endocardium and epicardium are segmented by propagating a reference contour over all the frames of the RT-CMR SAX scans. Our method was evaluated on 19 subjects, the accuracy of the automatic LV endocardium and epicardium segmentation was compared to those defined manually. The proposed method obtained a dice similarity coefficient (DSC) of 0.94 and a mean surface-to-surface distance (MSD) measure of 0.89 ± 0.53 mm. Additionally, a number of automatically obtained clinical measures were compared to ground truth values. On average we obtained a Pearson’s correlation coefficient (R) of 0.94 (0.99–0.74).

AB - We present a workflow for processing real-time cardiac MR (RT-CMR) scans for segmenting the left ventricle (LV) on short-axis slices (SAX). Our method is based on image registration, where the LV endocardium and epicardium are segmented by propagating a reference contour over all the frames of the RT-CMR SAX scans. Our method was evaluated on 19 subjects, the accuracy of the automatic LV endocardium and epicardium segmentation was compared to those defined manually. The proposed method obtained a dice similarity coefficient (DSC) of 0.94 and a mean surface-to-surface distance (MSD) measure of 0.89 ± 0.53 mm. Additionally, a number of automatically obtained clinical measures were compared to ground truth values. On average we obtained a Pearson’s correlation coefficient (R) of 0.94 (0.99–0.74).

KW - Left ventricle

KW - Realtime-MR

KW - Registration

KW - Segmentation

KW - Semi-automatic

UR - http://www.scopus.com/inward/record.url?scp=85048657855&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-92258-4_6

DO - 10.1007/978-3-319-92258-4_6

M3 - Conference contribution

AN - SCOPUS:85048657855

SN - 978-3-319-92257-7

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

SP - 57

EP - 66

BT - Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings

A2 - Klein, Stefan

A2 - Staring, Marius

A2 - Durrleman, Stanley

A2 - Sommer, Stefan

PB - Springer

CY - Cham

T2 - WBIR 2018

Y2 - 28 June 2018 through 29 June 2018

ER -

Shahzad R, Fasshauer M, Lelieveldt BPF, Lotz J, van der Geest RJ. Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation. In Klein S, Staring M, Durrleman S, Sommer S, editors, Biomedical Image Registration - 8th International Workshop, WBIR 2018 - Proceedings. Cham: Springer. 2018. p. 57-66. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-92258-4_6

Semi-automated Processing of Real-Time CMR Scans for Left Ventricle Segmentation

Abstract

Publication series

Workshop

Keywords

Access to Document

Other files and links

Fingerprint

Cite this