Automatic segmentation and disease classification using cardiac cine MR images

Jelmer M. Wolterink; Tim Leiner; Max A. Viergever; Ivana Išgum

doi:10.1007/978-3-319-75541-0_11

Automatic segmentation and disease classification using cardiac cine MR images

Jelmer M. Wolterink^*, Tim Leiner, Max A. Viergever, Ivana Išgum

^*Corresponding author for this work

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

86 Citations (Scopus)

12 Downloads (Pure)

Abstract

Segmentation of the heart in cardiac cine MR is clinically used to quantify cardiac function. We propose a fully automatic method for segmentation and disease classification using cardiac cine MR images. A convolutional neural network (CNN) was designed to simultaneously segment the left ventricle (LV), right ventricle (RV) and myocardium in end-diastole (ED) and end-systole (ES) images. Features derived from the obtained segmentations were used in a Random Forest classifier to label patients as suffering from dilated cardiomyopathy, hypertrophic cardiomyopathy, heart failure following myocardial infarction, right ventricular abnormality, or no cardiac disease. The method was developed and evaluated using a balanced dataset containing images of 100 patients, which was provided in the MICCAI 2017 automated cardiac diagnosis challenge (ACDC). Segmentation and classification pipeline were evaluated in a four-fold stratified cross-validation. Average Dice scores between reference and automatically obtained segmentations were 0.94, 0.88 and 0.87 for the LV, RV and myocardium. The classifier assigned 91% of patients to the correct disease category. Segmentation and disease classification took 5 s per patient. The results of our study suggest that image-based diagnosis using cine MR cardiac scans can be performed automatically with high accuracy.

Original language	English
Title of host publication	Statistical Atlases and Computational Models of the Heart
Subtitle of host publication	ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers
Editors	Olivier Bernard, Pierre-Marc Jodoin, Xiahai Zhuang, Guang Yang, Alistair Young, Maxime Sermesant, Alain Lalande, Mihaela Pop
Place of Publication	Cham
Publisher	Springer
Pages	101-110
Number of pages	10
ISBN (Electronic)	978-3-319-75541-0
ISBN (Print)	978-3-319-75540-3
DOIs	https://doi.org/10.1007/978-3-319-75541-0_11
Publication status	Published - 1 Jan 2018
Externally published	Yes
Event	8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017 - Quebec City, Canada Duration: 10 Sept 2017 → 14 Sept 2017 Conference number: 8

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	10663
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017
Abbreviated title	STACOM
Country/Territory	Canada
City	Quebec City
Period	10/09/17 → 14/09/17

Keywords

Automatic diagnosis
Cardiac MR
Convolutional neural networks
Deep learning
Random forest

Access to Document

10.1007/978-3-319-75541-0_11

Cite this

Wolterink, J. M., Leiner, T., Viergever, M. A., & Išgum, I. (2018). Automatic segmentation and disease classification using cardiac cine MR images. In O. Bernard, P.-M. Jodoin, X. Zhuang, G. Yang, A. Young, M. Sermesant, A. Lalande, & M. Pop (Eds.), Statistical Atlases and Computational Models of the Heart: ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers (pp. 101-110). (Lecture Notes in Computer Science; Vol. 10663). Springer. https://doi.org/10.1007/978-3-319-75541-0_11

Wolterink, Jelmer M. ; Leiner, Tim ; Viergever, Max A. et al. / Automatic segmentation and disease classification using cardiac cine MR images. Statistical Atlases and Computational Models of the Heart: ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers. editor / Olivier Bernard ; Pierre-Marc Jodoin ; Xiahai Zhuang ; Guang Yang ; Alistair Young ; Maxime Sermesant ; Alain Lalande ; Mihaela Pop. Cham : Springer, 2018. pp. 101-110 (Lecture Notes in Computer Science).

@inproceedings{a05601bc851f4c9e8e3a89aef87256d8,

title = "Automatic segmentation and disease classification using cardiac cine MR images",

abstract = "Segmentation of the heart in cardiac cine MR is clinically used to quantify cardiac function. We propose a fully automatic method for segmentation and disease classification using cardiac cine MR images. A convolutional neural network (CNN) was designed to simultaneously segment the left ventricle (LV), right ventricle (RV) and myocardium in end-diastole (ED) and end-systole (ES) images. Features derived from the obtained segmentations were used in a Random Forest classifier to label patients as suffering from dilated cardiomyopathy, hypertrophic cardiomyopathy, heart failure following myocardial infarction, right ventricular abnormality, or no cardiac disease. The method was developed and evaluated using a balanced dataset containing images of 100 patients, which was provided in the MICCAI 2017 automated cardiac diagnosis challenge (ACDC). Segmentation and classification pipeline were evaluated in a four-fold stratified cross-validation. Average Dice scores between reference and automatically obtained segmentations were 0.94, 0.88 and 0.87 for the LV, RV and myocardium. The classifier assigned 91% of patients to the correct disease category. Segmentation and disease classification took 5 s per patient. The results of our study suggest that image-based diagnosis using cine MR cardiac scans can be performed automatically with high accuracy.",

keywords = "Automatic diagnosis, Cardiac MR, Convolutional neural networks, Deep learning, Random forest",

author = "Wolterink, {Jelmer M.} and Tim Leiner and Viergever, {Max A.} and Ivana I{\v s}gum",

year = "2018",

month = jan,

day = "1",

doi = "10.1007/978-3-319-75541-0_11",

language = "English",

isbn = "978-3-319-75540-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "101--110",

editor = "Olivier Bernard and Pierre-Marc Jodoin and Xiahai Zhuang and Guang Yang and Alistair Young and Maxime Sermesant and Alain Lalande and Mihaela Pop",

booktitle = "Statistical Atlases and Computational Models of the Heart",

address = "Germany",

note = "8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017, STACOM ; Conference date: 10-09-2017 Through 14-09-2017",

}

Wolterink, JM, Leiner, T, Viergever, MA & Išgum, I 2018, Automatic segmentation and disease classification using cardiac cine MR images. in O Bernard, P-M Jodoin, X Zhuang, G Yang, A Young, M Sermesant, A Lalande & M Pop (eds), Statistical Atlases and Computational Models of the Heart: ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers. Lecture Notes in Computer Science, vol. 10663, Springer, Cham, pp. 101-110, 8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017, Quebec City, Quebec, Canada, 10/09/17. https://doi.org/10.1007/978-3-319-75541-0_11

Automatic segmentation and disease classification using cardiac cine MR images. / Wolterink, Jelmer M.; Leiner, Tim; Viergever, Max A. et al.
Statistical Atlases and Computational Models of the Heart: ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers. ed. / Olivier Bernard; Pierre-Marc Jodoin; Xiahai Zhuang; Guang Yang; Alistair Young; Maxime Sermesant; Alain Lalande; Mihaela Pop. Cham: Springer, 2018. p. 101-110 (Lecture Notes in Computer Science; Vol. 10663).

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

TY - GEN

T1 - Automatic segmentation and disease classification using cardiac cine MR images

AU - Wolterink, Jelmer M.

AU - Leiner, Tim

AU - Viergever, Max A.

AU - Išgum, Ivana

N1 - Conference code: 8

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Segmentation of the heart in cardiac cine MR is clinically used to quantify cardiac function. We propose a fully automatic method for segmentation and disease classification using cardiac cine MR images. A convolutional neural network (CNN) was designed to simultaneously segment the left ventricle (LV), right ventricle (RV) and myocardium in end-diastole (ED) and end-systole (ES) images. Features derived from the obtained segmentations were used in a Random Forest classifier to label patients as suffering from dilated cardiomyopathy, hypertrophic cardiomyopathy, heart failure following myocardial infarction, right ventricular abnormality, or no cardiac disease. The method was developed and evaluated using a balanced dataset containing images of 100 patients, which was provided in the MICCAI 2017 automated cardiac diagnosis challenge (ACDC). Segmentation and classification pipeline were evaluated in a four-fold stratified cross-validation. Average Dice scores between reference and automatically obtained segmentations were 0.94, 0.88 and 0.87 for the LV, RV and myocardium. The classifier assigned 91% of patients to the correct disease category. Segmentation and disease classification took 5 s per patient. The results of our study suggest that image-based diagnosis using cine MR cardiac scans can be performed automatically with high accuracy.

AB - Segmentation of the heart in cardiac cine MR is clinically used to quantify cardiac function. We propose a fully automatic method for segmentation and disease classification using cardiac cine MR images. A convolutional neural network (CNN) was designed to simultaneously segment the left ventricle (LV), right ventricle (RV) and myocardium in end-diastole (ED) and end-systole (ES) images. Features derived from the obtained segmentations were used in a Random Forest classifier to label patients as suffering from dilated cardiomyopathy, hypertrophic cardiomyopathy, heart failure following myocardial infarction, right ventricular abnormality, or no cardiac disease. The method was developed and evaluated using a balanced dataset containing images of 100 patients, which was provided in the MICCAI 2017 automated cardiac diagnosis challenge (ACDC). Segmentation and classification pipeline were evaluated in a four-fold stratified cross-validation. Average Dice scores between reference and automatically obtained segmentations were 0.94, 0.88 and 0.87 for the LV, RV and myocardium. The classifier assigned 91% of patients to the correct disease category. Segmentation and disease classification took 5 s per patient. The results of our study suggest that image-based diagnosis using cine MR cardiac scans can be performed automatically with high accuracy.

KW - Automatic diagnosis

KW - Cardiac MR

KW - Convolutional neural networks

KW - Deep learning

KW - Random forest

UR - http://www.scopus.com/inward/record.url?scp=85044445499&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-75541-0_11

DO - 10.1007/978-3-319-75541-0_11

M3 - Conference contribution

AN - SCOPUS:85044445499

SN - 978-3-319-75540-3

T3 - Lecture Notes in Computer Science

SP - 101

EP - 110

BT - Statistical Atlases and Computational Models of the Heart

A2 - Bernard, Olivier

A2 - Jodoin, Pierre-Marc

A2 - Zhuang, Xiahai

A2 - Yang, Guang

A2 - Young, Alistair

A2 - Sermesant, Maxime

A2 - Lalande, Alain

A2 - Pop, Mihaela

PB - Springer

CY - Cham

T2 - 8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017

Y2 - 10 September 2017 through 14 September 2017

ER -

Wolterink JM, Leiner T, Viergever MA, Išgum I. Automatic segmentation and disease classification using cardiac cine MR images. In Bernard O, Jodoin PM, Zhuang X, Yang G, Young A, Sermesant M, Lalande A, Pop M, editors, Statistical Atlases and Computational Models of the Heart: ACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers. Cham: Springer. 2018. p. 101-110. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-75541-0_11

Automatic segmentation and disease classification using cardiac cine MR images

Abstract

Publication series

Workshop

Keywords

Access to Document

Other files and links

Fingerprint

Cite this