Imperial College London
Portrait Picture

Dr Chen (Cherise) Chen

Faculty of Engineering, Department of Computing

Honorary Research Associate
 
 
 
//

Contact

 

chen.chen15 Website

 
 
//

Location

 

344Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@unpublished{Chen:2019,
author = {Chen, C and Bai, W and Davies, RH and Bhuva, AN and Manisty, C and Moon, JC and Aung, N and Lee, AM and Sanghvi, MM and Fung, K and Paiva, JM and Petersen, SE and Lukaschuk, E and Piechnik, SK and Neubauer, S and Rueckert, D},
publisher = {arXiv},
title = {Improving the generalizability of convolutional neural network-based segmentation on CMR images},
url = {https://arxiv.org/abs/1907.01268v2},
year = {2019}
}
Download

RIS format (EndNote, RefMan)

TY  - UNPB
AB  - Convolutional neural network (CNN) based segmentation methods provide anefficient and automated way for clinicians to assess the structure and functionof the heart in cardiac MR images. While CNNs can generally perform thesegmentation tasks with high accuracy when training and test images come fromthe same domain (e.g. same scanner or site), their performance often degradesdramatically on images from different scanners or clinical sites. We propose asimple yet effective way for improving the network generalization ability bycarefully designing data normalization and augmentation strategies toaccommodate common scenarios in multi-site, multi-scanner clinical imaging datasets. We demonstrate that a neural network trained on a single-sitesingle-scanner dataset from the UK Biobank can be successfully applied tosegmenting cardiac MR images across different sites and different scannerswithout substantial loss of accuracy. Specifically, the method was trained on alarge set of 3,975 subjects from the UK Biobank. It was then directly tested on600 different subjects from the UK Biobank for intra-domain testing and twoother sets for cross-domain testing: the ACDC dataset (100 subjects, 1 site, 2scanners) and the BSCMR-AS dataset (599 subjects, 6 sites, 9 scanners). Theproposed method produces promising segmentation results on the UK Biobank testset which are comparable to previously reported values in the literature, whilealso performing well on cross-domain test sets, achieving a mean Dice metric of0.90 for the left ventricle, 0.81 for the myocardium and 0.82 for the rightventricle on the ACDC dataset; and 0.89 for the left ventricle, 0.83 for themyocardium on the BSCMR-AS dataset. The proposed method offers a potentialsolution to improve CNN-based model generalizability for the cross-scanner andcross-site cardiac MR image segmentation task.
AU  - Chen,C
AU  - Bai,W
AU  - Davies,RH
AU  - Bhuva,AN
AU  - Manisty,C
AU  - Moon,JC
AU  - Aung,N
AU  - Lee,AM
AU  - Sanghvi,MM
AU  - Fung,K
AU  - Paiva,JM
AU  - Petersen,SE
AU  - Lukaschuk,E
AU  - Piechnik,SK
AU  - Neubauer,S
AU  - Rueckert,D
PB  - arXiv
PY  - 2019///
TI  - Improving the generalizability of convolutional neural network-based segmentation on CMR images
UR  - https://arxiv.org/abs/1907.01268v2
UR  - http://hdl.handle.net/10044/1/72573
ER  -
Download