Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension

Jinming Duan; Jo Schlemper; Wenjia Bai; Timothy J W Dawes; Ghalib Bello; Georgia Doumou; Antonio De Marvao; Declan P. O'Regan; Daniel Rueckert

Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension

Jinming Duan^*, Jo Schlemper, Wenjia Bai, Timothy J W Dawes, Ghalib Bello, Georgia Doumou, Antonio De Marvao, Declan P. O'Regan, Daniel Rueckert

^*Corresponding author for this work

Division of Informatics, Imaging & Data Sciences (L5)

Imperial College London

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

Abstract

In this paper we introduce a novel and accurate optimisation method for segmentation of cardiac MR (CMR) images in patients with pulmonary hypertension (PH). The proposed method explicitly takes into account the image features learned from a deep neural network. To this end, we estimate simultaneous probability maps over region and edge locations in CMR images using a fully convolutional network. Due to the distinct morphology of the heart in patients with PH, these probability maps can then be incorporated in a single nested level set optimisation framework to achieve multi-region segmentation with high efficiency. The proposed method uses an automatic way for level set initialisation and thus the whole optimisation is fully automated. We demonstrate that the proposed deep nested level set (DNLS) method outperforms existing state-of-the-art methods for CMR segmentation in PH patients.

Original language	English
Title of host publication	Medical Image Computing and Computer Assisted Intervention – MICCAI 2018
Subtitle of host publication	21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV
Editors	Alejandro F. Frangi, Julia A. Schnabel, Christos Davatzikos, Carlos Alberola-López, Gabor Fichtinger
Place of Publication	Cham
Publisher	Springer Cham
Pages	595–603
Number of pages	9
ISBN (Electronic)	9783030009373
ISBN (Print)	9783030009366
Publication status	Published - 14 Sept 2018

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	11073
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Cite this

Duan, J., Schlemper, J., Bai, W., J W Dawes, T., Bello, G., Doumou, G., Marvao, A. D., O'Regan, D. P., & Rueckert, D. (2018). Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension. In A. F. Frangi, J. A. Schnabel, C. Davatzikos, C. Alberola-López, & G. Fichtinger (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV (pp. 595–603). (Lecture Notes in Computer Science; Vol. 11073). Springer Cham.

Duan, Jinming ; Schlemper, Jo ; Bai, Wenjia et al. / Deep Nested Level Sets : Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV. editor / Alejandro F. Frangi ; Julia A. Schnabel ; Christos Davatzikos ; Carlos Alberola-López ; Gabor Fichtinger. Cham : Springer Cham, 2018. pp. 595–603 (Lecture Notes in Computer Science).

@inproceedings{52f8457a97e748ddaa1f150ca00d0601,

title = "Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension",

abstract = "In this paper we introduce a novel and accurate optimisation method for segmentation of cardiac MR (CMR) images in patients with pulmonary hypertension (PH). The proposed method explicitly takes into account the image features learned from a deep neural network. To this end, we estimate simultaneous probability maps over region and edge locations in CMR images using a fully convolutional network. Due to the distinct morphology of the heart in patients with PH, these probability maps can then be incorporated in a single nested level set optimisation framework to achieve multi-region segmentation with high efficiency. The proposed method uses an automatic way for level set initialisation and thus the whole optimisation is fully automated. We demonstrate that the proposed deep nested level set (DNLS) method outperforms existing state-of-the-art methods for CMR segmentation in PH patients.",

author = "Jinming Duan and Jo Schlemper and Wenjia Bai and {J W Dawes}, Timothy and Ghalib Bello and Georgia Doumou and Marvao, {Antonio De} and O'Regan, {Declan P.} and Daniel Rueckert",

year = "2018",

month = sep,

day = "14",

language = "English",

isbn = "9783030009366",

series = "Lecture Notes in Computer Science",

publisher = "Springer Cham",

pages = "595–603",

editor = "Frangi, {Alejandro F.} and Schnabel, {Julia A.} and Christos Davatzikos and Carlos Alberola-L{\'o}pez and Gabor Fichtinger",

booktitle = "Medical Image Computing and Computer Assisted Intervention – MICCAI 2018",

address = "Switzerland",

}

Duan, J, Schlemper, J, Bai, W, J W Dawes, T, Bello, G, Doumou, G, Marvao, AD, O'Regan, DP & Rueckert, D 2018, Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension. in AF Frangi, JA Schnabel, C Davatzikos, C Alberola-López & G Fichtinger (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV. Lecture Notes in Computer Science, vol. 11073, Springer Cham, Cham, pp. 595–603.

Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension. / Duan, Jinming; Schlemper, Jo; Bai, Wenjia et al.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV. ed. / Alejandro F. Frangi; Julia A. Schnabel; Christos Davatzikos; Carlos Alberola-López; Gabor Fichtinger. Cham: Springer Cham, 2018. p. 595–603 (Lecture Notes in Computer Science; Vol. 11073).

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

TY - GEN

T1 - Deep Nested Level Sets

T2 - Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension

AU - Duan, Jinming

AU - Schlemper, Jo

AU - Bai, Wenjia

AU - J W Dawes, Timothy

AU - Bello, Ghalib

AU - Doumou, Georgia

AU - Marvao, Antonio De

AU - O'Regan, Declan P.

AU - Rueckert, Daniel

PY - 2018/9/14

Y1 - 2018/9/14

N2 - In this paper we introduce a novel and accurate optimisation method for segmentation of cardiac MR (CMR) images in patients with pulmonary hypertension (PH). The proposed method explicitly takes into account the image features learned from a deep neural network. To this end, we estimate simultaneous probability maps over region and edge locations in CMR images using a fully convolutional network. Due to the distinct morphology of the heart in patients with PH, these probability maps can then be incorporated in a single nested level set optimisation framework to achieve multi-region segmentation with high efficiency. The proposed method uses an automatic way for level set initialisation and thus the whole optimisation is fully automated. We demonstrate that the proposed deep nested level set (DNLS) method outperforms existing state-of-the-art methods for CMR segmentation in PH patients.

AB - In this paper we introduce a novel and accurate optimisation method for segmentation of cardiac MR (CMR) images in patients with pulmonary hypertension (PH). The proposed method explicitly takes into account the image features learned from a deep neural network. To this end, we estimate simultaneous probability maps over region and edge locations in CMR images using a fully convolutional network. Due to the distinct morphology of the heart in patients with PH, these probability maps can then be incorporated in a single nested level set optimisation framework to achieve multi-region segmentation with high efficiency. The proposed method uses an automatic way for level set initialisation and thus the whole optimisation is fully automated. We demonstrate that the proposed deep nested level set (DNLS) method outperforms existing state-of-the-art methods for CMR segmentation in PH patients.

M3 - Conference contribution

SN - 9783030009366

T3 - Lecture Notes in Computer Science

SP - 595

EP - 603

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

A2 - Frangi, Alejandro F.

A2 - Schnabel, Julia A.

A2 - Davatzikos, Christos

A2 - Alberola-López, Carlos

A2 - Fichtinger, Gabor

PB - Springer Cham

CY - Cham

ER -

Duan J, Schlemper J, Bai W, J W Dawes T, Bello G, Doumou G et al. Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension. In Frangi AF, Schnabel JA, Davatzikos C, Alberola-López C, Fichtinger G, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part IV. Cham: Springer Cham. 2018. p. 595–603. (Lecture Notes in Computer Science). Epub 2018 Sept 13.

Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension

Abstract

Publication series

UN SDGs

Fingerprint

Cite this