Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training

Huaqi Qiu; Chen Qin; Loic Le Folgoc; Benjamin Hou; Jo Schlemper; Daniel Rueckert

doi:10.1007/978-3-030-39074-7_20

Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training

Huaqi Qiu
, Chen Qin
, Loic Le Folgoc
, Benjamin Hou
, Jo Schlemper
, Daniel Rueckert

Imperial College London

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

15 Citations (Scopus)

Résumé

Deep learning based registration methods have emerged as alternatives to traditional registration methods, with competitive accuracy and significantly less runtime. Two different strategies have been proposed to train such deep learning registration networks: supervised training strategy where the model is trained to regress to generated ground truth deformation; and unsupervised training strategy where the model directly optimises the similarity between the registered images. In this work, we directly compare the performance of these two training strategies for cardiac motion estimation on cardiac cine MR sequences. Testing on real cardiac MRI data shows that while the supervised training yields more regular deformation, the unsupervised more accurately captures the deformation of anatomical structures in cardiac motion.

langue originale	Anglais
titre	Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers
rédacteurs en chef	Mihaela Pop, Maxime Sermesant, Oscar Camara, Xiahai Zhuang, Shuo Li, Alistair Young, Tommaso Mansi, Avan Suinesiaputra
Editeur	Springer
Pages	186-194
Nombre de pages	9
ISBN (imprimé)	9783030390730
Les DOIs	https://doi.org/10.1007/978-3-030-39074-7_20
état	Publié - 1 janv. 2020
Modification externe	Oui
Evénement	10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 - Shenzhen, Chine Durée: 13 oct. 2019 → 13 oct. 2019

Série de publications

Nom	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12009 LNCS
ISSN (imprimé)	0302-9743
ISSN (Electronique)	1611-3349

Une conférence

Une conférence	10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019
Pays/Territoire	Chine
La ville	Shenzhen
période	13/10/19 → 13/10/19

Accès au document

10.1007/978-3-030-39074-7_20

Contient cette citation

Qiu, H., Qin, C., Le Folgoc, L., Hou, B., Schlemper, J., & Rueckert, D. (2020). Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training. Dans M. Pop, M. Sermesant, O. Camara, X. Zhuang, S. Li, A. Young, T. Mansi, & A. Suinesiaputra (eds.), Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers (p. 186-194). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol 12009 LNCS). Springer. https://doi.org/10.1007/978-3-030-39074-7_20

Qiu, Huaqi ; Qin, Chen ; Le Folgoc, Loic et al. / Deep Learning for Cardiac Motion Estimation : Supervised vs. Unsupervised Training. Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. Editeur / Mihaela Pop ; Maxime Sermesant ; Oscar Camara ; Xiahai Zhuang ; Shuo Li ; Alistair Young ; Tommaso Mansi ; Avan Suinesiaputra. Springer, 2020. p. 186-194 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{3315fb328f8a464092e61900c542be9e,

title = "Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training",

abstract = "Deep learning based registration methods have emerged as alternatives to traditional registration methods, with competitive accuracy and significantly less runtime. Two different strategies have been proposed to train such deep learning registration networks: supervised training strategy where the model is trained to regress to generated ground truth deformation; and unsupervised training strategy where the model directly optimises the similarity between the registered images. In this work, we directly compare the performance of these two training strategies for cardiac motion estimation on cardiac cine MR sequences. Testing on real cardiac MRI data shows that while the supervised training yields more regular deformation, the unsupervised more accurately captures the deformation of anatomical structures in cardiac motion.",

author = "Huaqi Qiu and Chen Qin and \{Le Folgoc\}, Loic and Benjamin Hou and Jo Schlemper and Daniel Rueckert",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2020.; 10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 ; Conference date: 13-10-2019 Through 13-10-2019",

year = "2020",

month = jan,

day = "1",

doi = "10.1007/978-3-030-39074-7\_20",

language = "English",

isbn = "9783030390730",

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

publisher = "Springer",

pages = "186--194",

editor = "Mihaela Pop and Maxime Sermesant and Oscar Camara and Xiahai Zhuang and Shuo Li and Alistair Young and Tommaso Mansi and Avan Suinesiaputra",

booktitle = "Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers",

}

Qiu, H, Qin, C, Le Folgoc, L, Hou, B, Schlemper, J & Rueckert, D 2020, Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training. Dans M Pop, M Sermesant, O Camara, X Zhuang, S Li, A Young, T Mansi & A Suinesiaputra (eds), Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), VOL. 12009 LNCS, Springer, p. 186-194, 10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019, Shenzhen, Chine, 13/10/19. https://doi.org/10.1007/978-3-030-39074-7_20

Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training. / Qiu, Huaqi; Qin, Chen; Le Folgoc, Loic et al.
Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. Ed. / Mihaela Pop; Maxime Sermesant; Oscar Camara; Xiahai Zhuang; Shuo Li; Alistair Young; Tommaso Mansi; Avan Suinesiaputra. Springer, 2020. p. 186-194 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol 12009 LNCS).

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

TY - GEN

T1 - Deep Learning for Cardiac Motion Estimation

T2 - 10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019

AU - Qiu, Huaqi

AU - Qin, Chen

AU - Le Folgoc, Loic

AU - Hou, Benjamin

AU - Schlemper, Jo

AU - Rueckert, Daniel

N1 - Publisher Copyright: © Springer Nature Switzerland AG 2020.

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Deep learning based registration methods have emerged as alternatives to traditional registration methods, with competitive accuracy and significantly less runtime. Two different strategies have been proposed to train such deep learning registration networks: supervised training strategy where the model is trained to regress to generated ground truth deformation; and unsupervised training strategy where the model directly optimises the similarity between the registered images. In this work, we directly compare the performance of these two training strategies for cardiac motion estimation on cardiac cine MR sequences. Testing on real cardiac MRI data shows that while the supervised training yields more regular deformation, the unsupervised more accurately captures the deformation of anatomical structures in cardiac motion.

AB - Deep learning based registration methods have emerged as alternatives to traditional registration methods, with competitive accuracy and significantly less runtime. Two different strategies have been proposed to train such deep learning registration networks: supervised training strategy where the model is trained to regress to generated ground truth deformation; and unsupervised training strategy where the model directly optimises the similarity between the registered images. In this work, we directly compare the performance of these two training strategies for cardiac motion estimation on cardiac cine MR sequences. Testing on real cardiac MRI data shows that while the supervised training yields more regular deformation, the unsupervised more accurately captures the deformation of anatomical structures in cardiac motion.

U2 - 10.1007/978-3-030-39074-7_20

DO - 10.1007/978-3-030-39074-7_20

M3 - Conference contribution

AN - SCOPUS:85081923511

SN - 9783030390730

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

SP - 186

EP - 194

BT - Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers

A2 - Pop, Mihaela

A2 - Sermesant, Maxime

A2 - Camara, Oscar

A2 - Zhuang, Xiahai

A2 - Li, Shuo

A2 - Young, Alistair

A2 - Mansi, Tommaso

A2 - Suinesiaputra, Avan

PB - Springer

Y2 - 13 October 2019 through 13 October 2019

ER -

Qiu H, Qin C, Le Folgoc L, Hou B, Schlemper J, Rueckert D. Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training. Dans Pop M, Sermesant M, Camara O, Zhuang X, Li S, Young A, Mansi T, Suinesiaputra A, rédacteurs en chef, Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges - 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. Springer. 2020. p. 186-194. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-39074-7_20

Deep Learning for Cardiac Motion Estimation: Supervised vs. Unsupervised Training

Résumé

Série de publications

Une conférence

Accès au document

Empreinte digitale

Contient cette citation