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 proceedingConference contributionAcademicpeer-review

68 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 languageEnglish
Title of host publicationStatistical Atlases and Computational Models of the Heart
Subtitle of host publicationACDC and MMWHS Challenges - 8th International Workshop, STACOM 2017, Revised Selected Papers
EditorsOlivier Bernard, Pierre-Marc Jodoin, Xiahai Zhuang, Guang Yang, Alistair Young, Maxime Sermesant, Alain Lalande, Mihaela Pop
Place of PublicationCham
PublisherSpringer
Pages101-110
Number of pages10
ISBN (Electronic)978-3-319-75541-0
ISBN (Print)978-3-319-75540-3
DOIs
Publication statusPublished - 1 Jan 2018
Externally publishedYes
Event8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017 - Quebec City, Canada
Duration: 10 Sept 201714 Sept 2017
Conference number: 8

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume10663
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Workshop

Workshop8th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2017
Abbreviated titleSTACOM
Country/TerritoryCanada
CityQuebec City
Period10/09/1714/09/17

Keywords

  • Automatic diagnosis
  • Cardiac MR
  • Convolutional neural networks
  • Deep learning
  • Random forest

Cite this