Combining Deep Learning and Shape Priors for Bi-Ventricular Segmentation of Volumetric Cardiac Magnetic Resonance Images

Jinming Duan, Jo Schlemper, Wenjia Bai, Timothy J.W. Dawes, Ghalib Bello, Carlo Biffi, Georgia Doumou, Antonio De Marvao, Declan P. O’Regan, Daniel Rueckert

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

3 Zitate (Scopus)

Abstract

In this paper, we combine a network-based method with image registration to develop a shape-based bi-ventricular segmentation tool for short-axis cardiac magnetic resonance (CMR) volumetric images. The method first employs a fully convolutional network (FCN) to learn the segmentation task from manually labelled ground truth CMR volumes. However, due to the presence of image artefacts in the training dataset, the resulting FCN segmentation results are often imperfect. As such, we propose a second step to refine the FCN segmentation. This step involves performing a non-rigid registration with multiple high-resolution bi-ventricular atlases, allowing the explicit shape priors to be inferred. We validate the proposed approach on 1831 healthy subjects and 200 subjects with pulmonary hypertension. Numerical experiments on the two datasets demonstrate that our approach is capable of producing accurate, high-resolution and anatomically smooth bi-ventricular models, despite the artefacts in the input CMR volumes.

OriginalspracheEnglisch
TitelShape in Medical Imaging - International Workshop, ShapeMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
Redakteure/-innenHervé Lombaert, Beatriz Paniagua, Bernhard Egger, Marcel Lüthi, Martin Reuter, Christian Wachinger
Herausgeber (Verlag)Springer Verlag
Seiten258-267
Seitenumfang10
ISBN (Print)9783030047467
DOIs
PublikationsstatusVeröffentlicht - 2018
Extern publiziertJa
VeranstaltungInternational Workshop on Shape in Medical Imaging, ShapeMI 2018 held in conjunction with 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spanien
Dauer: 20 Sept. 201820 Sept. 2018

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band11167 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

KonferenzInternational Workshop on Shape in Medical Imaging, ShapeMI 2018 held in conjunction with 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
Land/GebietSpanien
OrtGranada
Zeitraum20/09/1820/09/18

Fingerprint

Untersuchen Sie die Forschungsthemen von „Combining Deep Learning and Shape Priors for Bi-Ventricular Segmentation of Volumetric Cardiac Magnetic Resonance Images“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren