Learning interpretable anatomical features through deep generative models: Application to cardiac remodeling
File(s)
Accepted version
Author(s)
Type
Conference Paper
Abstract
Alterations in the geometry and function of the heart define well-established causes of cardiovascular disease. However, current approaches to the diagnosis of cardiovascular diseases often rely on subjective human assessment as well as manual analysis of medical images. Both factors limit the sensitivity in quantifying complex structural and functional phenotypes. Deep learning approaches have recently achieved success for tasks such as classification or segmentation of medical images, but lack interpretability in the feature extraction and decision processes, limiting their value in clinical diagnosis. In this work, we propose a 3D convolutional generative model for automatic classification of images from patients with cardiac diseases associated with structural remodeling. The model leverages interpretable task-specific anatomic patterns learned from 3D segmentations. It further allows to visualise and quantify the learned pathology-specific remodeling patterns in the original input space of the images. This approach yields high accuracy in the categorization of healthy and hypertrophic cardiomyopathy subjects when tested on unseen MR images from our own multi-centre dataset (100%) as well on the ACDC MICCAI 2017 dataset (90%). We believe that the proposed deep learning approach is a promising step towards the development of interpretable classifiers for the medical imaging domain, which may help clinicians to improve diagnostic accuracy and enhance patient risk-stratification.
Date Issued
2018-09-16
Date Acceptance
2018-05-25
Citation
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2018, 11071 LNCS, pp.464-471
ISBN
9783030009335
ISSN
0302-9743
Publisher
Springer
Start Page
464
End Page
471
Journal / Book Title
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume
11071 LNCS
Copyright Statement
© Springer Nature Switzerland AG 2018. The final publication is available at Springer via https://link.springer.com/chapter/10.1007%2F978-3-030-00934-2_52
Sponsor
British Heart Foundation
The Academy of Medical Sciences
Imperial College Healthcare NHS Trust- BRC Funding
British Heart Foundation
Imperial College Healthcare NHS Trust- BRC Funding
Grant Number
RE/08/002/23906
nil
RDC04
NH/17/1/32725
RDB02
Source
International Conference On Medical Image Computing & Computer Assisted Intervention
Subjects
Science & Technology
Technology
Computer Science, Theory & Methods
Computer Science
TASK-FORCE
MANAGEMENT
cs.CV
cs.CV
08 Information and Computing Sciences
Artificial Intelligence & Image Processing
Publication Status
Published
Start Date
2018-09-16
Finish Date
2018-09-20
Coverage Spatial
Granada, Spain
Date Publish Online
2018-09-26