Domain-Adversarial Learning for Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac MR Image Segmentation

TY - CHAP

T1 - Domain-Adversarial Learning for Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac MR Image Segmentation

AU - Scannell, Cian M.

AU - Chiribiri, Amedeo

AU - Veta, Mitko

N1 - Funding Information: The authors of this paper declare that the segmentation method they implemented for participation in the M&Ms challenge has not used any pre-trained models nor additional MRI datasets other than those provided by the organizers. The authors acknowledge financial support from the King?s College London & Imperial College London EPSRC Centre for Doctoral Training in Medical Imaging (EP/L015226/1); Philips Healthcare; and The Centre of Excellence in Medical Engineering funded by the Wellcome Trust and EPSRC under grant number WT 203148/Z/16/Z. Publisher Copyright: © 2021, Springer Nature Switzerland AG. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021

Y1 - 2021

N2 - Cine cardiac magnetic resonance (CMR) has become the gold standard for the non-invasive evaluation of cardiac function. In particular, it allows the accurate quantification of functional parameters including the chamber volumes and ejection fraction. Deep learning has shown the potential to automate the requisite cardiac structure segmentation. However, the lack of robustness of deep learning models has hindered their widespread clinical adoption. Due to differences in the data characteristics, neural networks trained on data from a specific scanner are not guaranteed to generalise well to data acquired at a different centre or with a different scanner. In this work, we propose a principled solution to the problem of this domain shift. Domain-adversarial learning is used to train a domain-invariant 2D U-Net using labelled and unlabelled data. This approach is evaluated on both seen and unseen domains from the M&Ms challenge dataset and the domain-adversarial approach shows improved performance as compared to standard training. Additionally, we show that the domain information cannot be recovered from the learned features.

AB - Cine cardiac magnetic resonance (CMR) has become the gold standard for the non-invasive evaluation of cardiac function. In particular, it allows the accurate quantification of functional parameters including the chamber volumes and ejection fraction. Deep learning has shown the potential to automate the requisite cardiac structure segmentation. However, the lack of robustness of deep learning models has hindered their widespread clinical adoption. Due to differences in the data characteristics, neural networks trained on data from a specific scanner are not guaranteed to generalise well to data acquired at a different centre or with a different scanner. In this work, we propose a principled solution to the problem of this domain shift. Domain-adversarial learning is used to train a domain-invariant 2D U-Net using labelled and unlabelled data. This approach is evaluated on both seen and unseen domains from the M&Ms challenge dataset and the domain-adversarial approach shows improved performance as compared to standard training. Additionally, we show that the domain information cannot be recovered from the learned features.

KW - Cardiac MRI segmentation

KW - Domain-adversarial learning

UR - http://www.scopus.com/inward/record.url?scp=85101531547&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-68107-4_23

DO - 10.1007/978-3-030-68107-4_23

M3 - Conference paper

AN - SCOPUS:85101531547

SN - 9783030681067

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

SP - 228

EP - 237

BT - Statistical Atlases and Computational Models of the Heart. MandMs and EMIDEC Challenges - 11th International Workshop, STACOM 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers

A2 - Puyol Anton, Esther

A2 - Pop, Mihaela

A2 - Sermesant, Maxime

A2 - Campello, Victor

A2 - Lalande, Alain

A2 - Lekadir, Karim

A2 - Suinesiaputra, Avan

A2 - Camara, Oscar

A2 - Young, Alistair

PB - Springer Science and Business Media Deutschland GmbH

T2 - 11th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2020 held in Conjunction with MICCAI 2020

Y2 - 4 October 2020 through 4 October 2020

ER -