Improving the Generalizability of Convolutional Neural Network-Based Segmentation on CMR Images
Open Access
- 30 June 2020
- journal article
- research article
- Published by Frontiers Media SA in Frontiers in Cardiovascular Medicine
- Vol. 7, 105
Abstract
Background: Convolutional neural network (CNN) based segmentation methods provide an efficient and automated way for clinicians to assess the structure and function of the heart in cardiac MR images. While CNNs can generally perform the segmentation tasks with high accuracy when training and test images come from the same domain (e.g., same scanner or site), their performance often degrades dramatically on images from different scanners or clinical sites. Methods: We propose a simple yet effective way for improving the network generalization ability by carefully designing data normalization and augmentation strategies to accommodate common scenarios in multi-site, multi-scanner clinical imaging data sets. We demonstrate that a neural network trained on a single-site single-scanner dataset from the UK Biobank can be successfully applied to segmenting cardiac MR images across different sites and different scanners without substantial loss of accuracy. Specifically, the method was trained on a large set of 3,975 subjects from the UK Biobank. It was then directly tested on 600 different subjects from the UK Biobank for intra-domain testing and two other sets for cross-domain testing: the ACDC dataset (100 subjects, 1 site, 2 scanners) and the BSCMR-AS dataset (599 subjects, 6 sites, 9 scanners). Results: The proposed method produces promising segmentation results on the UK Biobank test set which are comparable to previously reported values in the literature, while also performing well on cross-domain test sets, achieving a mean Dice metric of 0.90 for the left ventricle, 0.81 for the myocardium, and 0.82 for the right ventricle on the ACDC dataset; and 0.89 for the left ventricle, 0.83 for the myocardium on the BSCMR-AS dataset. Conclusions: The proposed method offers a potential solution to improve CNN-based model generalizability for the cross-scanner and cross-site cardiac MR image segmentation task.Keywords
All Related Versions
This publication has 29 references indexed in Scilit:
- Current artefacts in cardiac and chest magnetic resonance imaging: tips and tricksBritish Journal of Radiology, 2016
- UK Biobank's cardiovascular magnetic resonance protocolJournal of Cardiovascular Magnetic Resonance, 2016
- The incremental impact of cardiac MRI on clinical decision-makingBritish Journal of Radiology, 2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Rotation-invariant convolutional neural networks for galaxy morphology predictionMonthly Notices of the Royal Astronomical Society, 2015
- Left ventricular shape variation in asymptomatic populations: the multi-ethnic study of atherosclerosisJournal of Cardiovascular Magnetic Resonance, 2014
- Cardiovascular MR Imaging at 3 T: Opportunities, Challenges, and SolutionsRadioGraphics, 2014
- A collaborative resource to build consensus for automated left ventricular segmentation of cardiac MR imagesMedical Image Analysis, 2014
- A review of segmentation methods in short axis cardiac MR imagesMedical Image Analysis, 2011
- Comprehensive Cardiac Magnetic Resonance Imaging at 3.0 TeslaInvestigative Radiology, 2006