UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets

Dorent, R; Booth, T; Li, W; Sudre, CH; Kafiabadi, S; Cardoso, J; Ourselin, S; (2021) Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets. Medical Image Analysis , 67 , Article 101862. 10.1016/j.media.2020.101862. Green open access

[thumbnail of Cardoso_1-s2.0-S1361841520302267-main.pdf]
Preview
Text
Cardoso_1-s2.0-S1361841520302267-main.pdf - Published Version

Download (4MB) | Preview

Abstract

Brain tissue segmentation from multimodal MRI is a key building block of many neuroimaging analysis pipelines. Established tissue segmentation approaches have, however, not been developed to cope with large anatomical changes resulting from pathology, such as white matter lesions or tumours, and often fail in these cases. In the meantime, with the advent of deep neural networks (DNNs), segmentation of brain lesions has matured significantly. However, few existing approaches allow for the joint segmentation of normal tissue and brain lesions. Developing a DNN for such a joint task is currently hampered by the fact that annotated datasets typically address only one specific task and rely on task-specific imaging protocols including a task-specific set of imaging modalities. In this work, we propose a novel approach to build a joint tissue and lesion segmentation model from aggregated task-specific hetero-modal domain-shifted and partially-annotated datasets. Starting from a variational formulation of the joint problem, we show how the expected risk can be decomposed and optimised empirically. We exploit an upper bound of the risk to deal with heterogeneous imaging modalities across datasets. To deal with potential domain shift, we integrated and tested three conventional techniques based on data augmentation, adversarial learning and pseudo-healthy generation. For each individual task, our joint approach reaches comparable performance to task-specific and fully-supervised models. The proposed framework is assessed on two different types of brain lesions: White matter lesions and gliomas. In the latter case, lacking a joint ground-truth for quantitative assessment purposes, we propose and use a novel clinically-relevant qualitative assessment methodology.

Type: Article
Title: Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.media.2020.101862
Publisher version: https://doi.org/10.1016/j.media.2020.101862
Language: English
Additional information: Copyright © 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Keywords: Joint Learning, Domain Adaptation, Multi-Task Learning, Multi-Modal
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Population Health Sciences > Institute of Cardiovascular Science
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Population Health Sciences > Institute of Cardiovascular Science > Population Science and Experimental Medicine
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Population Health Sciences > Institute of Cardiovascular Science > Population Science and Experimental Medicine > MRC Unit for Lifelong Hlth and Ageing
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng
URI: https://discovery.ucl.ac.uk/id/eprint/10126717
Downloads since deposit
26Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item