Imperial College London

Professor Paul M. Matthews

Faculty of MedicineDepartment of Brain Sciences

Edmond and Lily Safra Chair, Head of Department



+44 (0)20 7594 2855p.matthews




Ms Siobhan Dillon +44 (0)20 7594 2855




E502Burlington DanesHammersmith Campus






BibTex format

author = {Bai, W and Sinclair, M and Tarroni, G and Oktay, O and Rajchl, M and Vaillant, G and Lee, AM and Aung, N and Lukaschuk, E and Sanghvi, MM and Zemrak, F and Fung, K and Paiva, JM and Carapella, V and Kim, YJ and Suzuki, H and Kainz, B and Matthews, PM and Petersen, SE and Piechnik, SK and Neubauer, S and Glocker, B and Rueckert, D},
doi = {10.1186/s12968-018-0471-x},
journal = {Journal of Cardiovascular Magnetic Resonance},
title = {Automated cardiovascular magnetic resonance image analysis with fully convolutional networks.},
url = {},
volume = {20},
year = {2018}

RIS format (EndNote, RefMan)

AB - Background: Cardiovascular magnetic resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR imageanalysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images.Methods: Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV)end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV).Results: By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. On a short-axis image test set of 600 subjects, it achieves an average Dice metric of 0.94 for the LV cavity, 0.88 for the LV myocardium and 0.90 for the RV cavity. The meanabsolute difference between automated measurement and manual measurement was 6.1 mL for LVEDV, 5.3 mL for LVESV, 6.9 gram for LVM, 8.5 mL for RVEDV and 7.2 mL for RVESV. On long-ax
AU - Bai,W
AU - Sinclair,M
AU - Tarroni,G
AU - Oktay,O
AU - Rajchl,M
AU - Vaillant,G
AU - Lee,AM
AU - Aung,N
AU - Lukaschuk,E
AU - Sanghvi,MM
AU - Zemrak,F
AU - Fung,K
AU - Paiva,JM
AU - Carapella,V
AU - Kim,YJ
AU - Suzuki,H
AU - Kainz,B
AU - Matthews,PM
AU - Petersen,SE
AU - Piechnik,SK
AU - Neubauer,S
AU - Glocker,B
AU - Rueckert,D
DO - 10.1186/s12968-018-0471-x
PY - 2018///
SN - 1097-6647
TI - Automated cardiovascular magnetic resonance image analysis with fully convolutional networks.
T2 - Journal of Cardiovascular Magnetic Resonance
UR -
UR -
VL - 20
ER -