Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images

Sana Jabbar; Syed Talha Bukhari; Hassan Mohy-ud-Din

doi:10.1007/978-3-030-93722-5_30

Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images

Sana Jabbar
, Syed Talha Bukhari
, Hassan Mohy-ud-Din^*

^*Corresponding author for this work

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

5 Scopus citations

Abstract

We proposed a multi-view SA-LA model for simultaneous segmentation of RV on the short-axis (SA) and long-axis (LA) cardiac MR images. The multi-view SA-LA model is a multi-encoder, multi-decoder U-Net architecture based on the U-Net model. One encoder-decoder pair segments the RV on SA images and the other pair on LA images. Multi-view SA-LA model assembles an extremely rich set of synergistic features, at the root of the encoder branch, by combining feature maps learned from matched SA and LA cardiac MR images. Segmentation performance is further enhanced by: (1) incorporating spatial context of LV as a prior and (2) performing deep supervision in the last three layers of the decoder branch. Multi-view SA-LA model was extensively evaluated on the MICCAI 2021 Multi- Disease, Multi-View, and Multi- Centre RV Segmentation Challenge dataset (M&Ms-2021). M&Ms-2021 dataset consists of multi-phase, multi-view cardiac MR images of 360 subjects acquired at four clinical centers with three different vendors. On the challenge cohort (160 subjects), the proposed multi-view SA-LA model achieved a Dice Score of 91 % and Hausdorff distance of 11.2 mm on short-axis images and a Dice Score of 89.6 % and Hausdorff distance of 8.1 mm on long-axis images. Moreover, multi-view SA-LA model exhibited strong generalization to unseen RV related pathologies including Dilated Right Ventricle (DSC: SA 91.41 %, LA 89.63 % ) and Tricuspidal Regurgitation (DSC: SA 91.40 %, LA 90.40 % ) with low variance (σ_DSC : SA < 5 %, LA < 6 % ).

Original language	English
Title of host publication	Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers
Editors	Esther Puyol Antón, Alistair Young, Avan Suinesiaputra, Mihaela Pop, Carlos Martín-Isla, Maxime Sermesant, Oscar Camara, Karim Lekadir
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	277-286
Number of pages	10
ISBN (Print)	9783030937218
DOIs	https://doi.org/10.1007/978-3-030-93722-5_30
State	Published - 2022
Externally published	Yes
Event	12th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2021 held in conjunction with MICCAI 2021 - Strasbourg, France Duration: 27 Sep 2021 → 27 Sep 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13131 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	12th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2021 held in conjunction with MICCAI 2021
Country/Territory	France
City	Strasbourg
Period	27/09/21 → 27/09/21

Bibliographical note

Publisher Copyright:
© 2022, Springer Nature Switzerland AG.

Keywords

Cardiac imaging
Convolutional Neural Network
Deep neural network
Long-axis sequence
Magnetic resonance imaging
Right Ventricle
Segmentation
Short-axis sequence
U-Net

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-93722-5_30

Cite this

Jabbar, S., Bukhari, S. T., & Mohy-ud-Din, H. (2022). Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images. In E. Puyol Antón, A. Young, A. Suinesiaputra, M. Pop, C. Martín-Isla, M. Sermesant, O. Camara, & K. Lekadir (Eds.), Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers (pp. 277-286). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13131 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-93722-5_30

Jabbar, Sana ; Bukhari, Syed Talha ; Mohy-ud-Din, Hassan. / Multi-view SA-LA Net : A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images. Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. editor / Esther Puyol Antón ; Alistair Young ; Avan Suinesiaputra ; Mihaela Pop ; Carlos Martín-Isla ; Maxime Sermesant ; Oscar Camara ; Karim Lekadir. Springer Science and Business Media Deutschland GmbH, 2022. pp. 277-286 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ab3aa29d6dcd499790c3edd89d066d1a,

title = "Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images",

abstract = "We proposed a multi-view SA-LA model for simultaneous segmentation of RV on the short-axis (SA) and long-axis (LA) cardiac MR images. The multi-view SA-LA model is a multi-encoder, multi-decoder U-Net architecture based on the U-Net model. One encoder-decoder pair segments the RV on SA images and the other pair on LA images. Multi-view SA-LA model assembles an extremely rich set of synergistic features, at the root of the encoder branch, by combining feature maps learned from matched SA and LA cardiac MR images. Segmentation performance is further enhanced by: (1) incorporating spatial context of LV as a prior and (2) performing deep supervision in the last three layers of the decoder branch. Multi-view SA-LA model was extensively evaluated on the MICCAI 2021 Multi- Disease, Multi-View, and Multi- Centre RV Segmentation Challenge dataset (M\&Ms-2021). M\&Ms-2021 dataset consists of multi-phase, multi-view cardiac MR images of 360 subjects acquired at four clinical centers with three different vendors. On the challenge cohort (160 subjects), the proposed multi-view SA-LA model achieved a Dice Score of 91 \% and Hausdorff distance of 11.2 mm on short-axis images and a Dice Score of 89.6 \% and Hausdorff distance of 8.1 mm on long-axis images. Moreover, multi-view SA-LA model exhibited strong generalization to unseen RV related pathologies including Dilated Right Ventricle (DSC: SA 91.41 \%, LA 89.63 \% ) and Tricuspidal Regurgitation (DSC: SA 91.40 \%, LA 90.40 \% ) with low variance (σDSC : SA < 5 \%, LA < 6 \% ).",

keywords = "Cardiac imaging, Convolutional Neural Network, Deep neural network, Long-axis sequence, Magnetic resonance imaging, Right Ventricle, Segmentation, Short-axis sequence, U-Net",

author = "Sana Jabbar and Bukhari, \{Syed Talha\} and Hassan Mohy-ud-Din",

note = "Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.; 12th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2021 held in conjunction with MICCAI 2021 ; Conference date: 27-09-2021 Through 27-09-2021",

year = "2022",

doi = "10.1007/978-3-030-93722-5\_30",

language = "English",

isbn = "9783030937218",

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

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "277--286",

editor = "\{Puyol Ant{\'o}n\}, Esther and Alistair Young and Avan Suinesiaputra and Mihaela Pop and Carlos Mart{\'i}n-Isla and Maxime Sermesant and Oscar Camara and Karim Lekadir",

booktitle = "Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers",

address = "Germany",

}

Jabbar, S, Bukhari, ST & Mohy-ud-Din, H 2022, Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images. in E Puyol Antón, A Young, A Suinesiaputra, M Pop, C Martín-Isla, M Sermesant, O Camara & K Lekadir (eds), Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13131 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 277-286, 12th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2021 held in conjunction with MICCAI 2021, Strasbourg, France, 27/09/21. https://doi.org/10.1007/978-3-030-93722-5_30

Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images. / Jabbar, Sana; Bukhari, Syed Talha; Mohy-ud-Din, Hassan.
Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. ed. / Esther Puyol Antón; Alistair Young; Avan Suinesiaputra; Mihaela Pop; Carlos Martín-Isla; Maxime Sermesant; Oscar Camara; Karim Lekadir. Springer Science and Business Media Deutschland GmbH, 2022. p. 277-286 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13131 LNCS).

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

TY - GEN

T1 - Multi-view SA-LA Net

T2 - 12th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2021 held in conjunction with MICCAI 2021

AU - Jabbar, Sana

AU - Bukhari, Syed Talha

AU - Mohy-ud-Din, Hassan

PY - 2022

Y1 - 2022

N2 - We proposed a multi-view SA-LA model for simultaneous segmentation of RV on the short-axis (SA) and long-axis (LA) cardiac MR images. The multi-view SA-LA model is a multi-encoder, multi-decoder U-Net architecture based on the U-Net model. One encoder-decoder pair segments the RV on SA images and the other pair on LA images. Multi-view SA-LA model assembles an extremely rich set of synergistic features, at the root of the encoder branch, by combining feature maps learned from matched SA and LA cardiac MR images. Segmentation performance is further enhanced by: (1) incorporating spatial context of LV as a prior and (2) performing deep supervision in the last three layers of the decoder branch. Multi-view SA-LA model was extensively evaluated on the MICCAI 2021 Multi- Disease, Multi-View, and Multi- Centre RV Segmentation Challenge dataset (M&Ms-2021). M&Ms-2021 dataset consists of multi-phase, multi-view cardiac MR images of 360 subjects acquired at four clinical centers with three different vendors. On the challenge cohort (160 subjects), the proposed multi-view SA-LA model achieved a Dice Score of 91 % and Hausdorff distance of 11.2 mm on short-axis images and a Dice Score of 89.6 % and Hausdorff distance of 8.1 mm on long-axis images. Moreover, multi-view SA-LA model exhibited strong generalization to unseen RV related pathologies including Dilated Right Ventricle (DSC: SA 91.41 %, LA 89.63 % ) and Tricuspidal Regurgitation (DSC: SA 91.40 %, LA 90.40 % ) with low variance (σDSC : SA < 5 %, LA < 6 % ).

AB - We proposed a multi-view SA-LA model for simultaneous segmentation of RV on the short-axis (SA) and long-axis (LA) cardiac MR images. The multi-view SA-LA model is a multi-encoder, multi-decoder U-Net architecture based on the U-Net model. One encoder-decoder pair segments the RV on SA images and the other pair on LA images. Multi-view SA-LA model assembles an extremely rich set of synergistic features, at the root of the encoder branch, by combining feature maps learned from matched SA and LA cardiac MR images. Segmentation performance is further enhanced by: (1) incorporating spatial context of LV as a prior and (2) performing deep supervision in the last three layers of the decoder branch. Multi-view SA-LA model was extensively evaluated on the MICCAI 2021 Multi- Disease, Multi-View, and Multi- Centre RV Segmentation Challenge dataset (M&Ms-2021). M&Ms-2021 dataset consists of multi-phase, multi-view cardiac MR images of 360 subjects acquired at four clinical centers with three different vendors. On the challenge cohort (160 subjects), the proposed multi-view SA-LA model achieved a Dice Score of 91 % and Hausdorff distance of 11.2 mm on short-axis images and a Dice Score of 89.6 % and Hausdorff distance of 8.1 mm on long-axis images. Moreover, multi-view SA-LA model exhibited strong generalization to unseen RV related pathologies including Dilated Right Ventricle (DSC: SA 91.41 %, LA 89.63 % ) and Tricuspidal Regurgitation (DSC: SA 91.40 %, LA 90.40 % ) with low variance (σDSC : SA < 5 %, LA < 6 % ).

KW - Cardiac imaging

KW - Convolutional Neural Network

KW - Deep neural network

KW - Long-axis sequence

KW - Magnetic resonance imaging

KW - Right Ventricle

KW - Segmentation

KW - Short-axis sequence

KW - U-Net

UR - https://www.scopus.com/pages/publications/85124006382

U2 - 10.1007/978-3-030-93722-5_30

DO - 10.1007/978-3-030-93722-5_30

M3 - Conference contribution

AN - SCOPUS:85124006382

SN - 9783030937218

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

SP - 277

EP - 286

BT - Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers

A2 - Puyol Antón, Esther

A2 - Young, Alistair

A2 - Suinesiaputra, Avan

A2 - Pop, Mihaela

A2 - Martín-Isla, Carlos

A2 - Sermesant, Maxime

A2 - Camara, Oscar

A2 - Lekadir, Karim

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 27 September 2021 through 27 September 2021

ER -

Jabbar S, Bukhari ST, Mohy-ud-Din H. Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images. In Puyol Antón E, Young A, Suinesiaputra A, Pop M, Martín-Isla C, Sermesant M, Camara O, Lekadir K, editors, Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge - 12th International Workshop, STACOM 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 277-286. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-93722-5_30

Multi-view SA-LA Net: A Framework for Simultaneous Segmentation of RV on Multi-view Cardiac MR Images

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this