King's College London

Research portal

Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework

Research output: Chapter in Book/Report/Conference proceedingConference paper

Standard

Combining Multimodal Information for Metal Artefact Reduction : An Unsupervised Deep Learning Framework. / Ranzini, Marta B.M.; Groothuis, Irme; Klaser, Kerstin; Cardoso, M. Jorge; Henckel, Johann; Ourselin, Sebastien; Hart, Alister; Modat, Marc.

ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2020. p. 600-604 9098633 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2020-April).

Research output: Chapter in Book/Report/Conference proceedingConference paper

Harvard

Ranzini, MBM, Groothuis, I, Klaser, K, Cardoso, MJ, Henckel, J, Ourselin, S, Hart, A & Modat, M 2020, Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework. in ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging., 9098633, Proceedings - International Symposium on Biomedical Imaging, vol. 2020-April, IEEE Computer Society, pp. 600-604, 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020, Iowa City, United States, 3/04/2020. https://doi.org/10.1109/ISBI45749.2020.9098633

APA

Ranzini, M. B. M., Groothuis, I., Klaser, K., Cardoso, M. J., Henckel, J., Ourselin, S., Hart, A., & Modat, M. (2020). Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework. In ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging (pp. 600-604). [9098633] (Proceedings - International Symposium on Biomedical Imaging; Vol. 2020-April). IEEE Computer Society. https://doi.org/10.1109/ISBI45749.2020.9098633

Vancouver

Ranzini MBM, Groothuis I, Klaser K, Cardoso MJ, Henckel J, Ourselin S et al. Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework. In ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society. 2020. p. 600-604. 9098633. (Proceedings - International Symposium on Biomedical Imaging). https://doi.org/10.1109/ISBI45749.2020.9098633

Author

Ranzini, Marta B.M. ; Groothuis, Irme ; Klaser, Kerstin ; Cardoso, M. Jorge ; Henckel, Johann ; Ourselin, Sebastien ; Hart, Alister ; Modat, Marc. / Combining Multimodal Information for Metal Artefact Reduction : An Unsupervised Deep Learning Framework. ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2020. pp. 600-604 (Proceedings - International Symposium on Biomedical Imaging).

Bibtex Download

@inbook{a266837f0eca41ecbe36e92c847501f4,
title = "Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework",
abstract = "Metal artefact reduction (MAR) techniques aim at removing metal-induced noise from clinical images. In Computed Tomography (CT), supervised deep learning approaches have been shown effective but limited in generalisability, as they mostly rely on synthetic data. In Magnetic Resonance Imaging (MRI) instead, no method has yet been introduced to correct the susceptibility artefact, still present even in MAR-specific acquisitions. In this work, we hypothesise that a multimodal approach to MAR would improve both CT and MRI. Given their different artefact appearance, their complementary information can compensate for the corrupted signal in either modality. We thus propose an unsupervised deep learning method for multimodal MAR. We introduce the use of Locally Normalised Cross Correlation as a loss term to encourage the fusion of multimodal information. Experiments show that our approach favours a smoother correction in the CT, while promoting signal recovery in the MRI.",
keywords = "CT, Deep Learning, Metal Artefact Reduction, MR, Unsupervised Learning",
author = "Ranzini, {Marta B.M.} and Irme Groothuis and Kerstin Klaser and Cardoso, {M. Jorge} and Johann Henckel and Sebastien Ourselin and Alister Hart and Marc Modat",
year = "2020",
month = apr,
doi = "10.1109/ISBI45749.2020.9098633",
language = "English",
series = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",
pages = "600--604",
booktitle = "ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging",
note = "17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 ; Conference date: 03-04-2020 Through 07-04-2020",

}

RIS (suitable for import to EndNote) Download

TY - CHAP

T1 - Combining Multimodal Information for Metal Artefact Reduction

T2 - 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020

AU - Ranzini, Marta B.M.

AU - Groothuis, Irme

AU - Klaser, Kerstin

AU - Cardoso, M. Jorge

AU - Henckel, Johann

AU - Ourselin, Sebastien

AU - Hart, Alister

AU - Modat, Marc

PY - 2020/4

Y1 - 2020/4

N2 - Metal artefact reduction (MAR) techniques aim at removing metal-induced noise from clinical images. In Computed Tomography (CT), supervised deep learning approaches have been shown effective but limited in generalisability, as they mostly rely on synthetic data. In Magnetic Resonance Imaging (MRI) instead, no method has yet been introduced to correct the susceptibility artefact, still present even in MAR-specific acquisitions. In this work, we hypothesise that a multimodal approach to MAR would improve both CT and MRI. Given their different artefact appearance, their complementary information can compensate for the corrupted signal in either modality. We thus propose an unsupervised deep learning method for multimodal MAR. We introduce the use of Locally Normalised Cross Correlation as a loss term to encourage the fusion of multimodal information. Experiments show that our approach favours a smoother correction in the CT, while promoting signal recovery in the MRI.

AB - Metal artefact reduction (MAR) techniques aim at removing metal-induced noise from clinical images. In Computed Tomography (CT), supervised deep learning approaches have been shown effective but limited in generalisability, as they mostly rely on synthetic data. In Magnetic Resonance Imaging (MRI) instead, no method has yet been introduced to correct the susceptibility artefact, still present even in MAR-specific acquisitions. In this work, we hypothesise that a multimodal approach to MAR would improve both CT and MRI. Given their different artefact appearance, their complementary information can compensate for the corrupted signal in either modality. We thus propose an unsupervised deep learning method for multimodal MAR. We introduce the use of Locally Normalised Cross Correlation as a loss term to encourage the fusion of multimodal information. Experiments show that our approach favours a smoother correction in the CT, while promoting signal recovery in the MRI.

KW - CT

KW - Deep Learning

KW - Metal Artefact Reduction

KW - MR

KW - Unsupervised Learning

UR - http://www.scopus.com/inward/record.url?scp=85085863009&partnerID=8YFLogxK

U2 - 10.1109/ISBI45749.2020.9098633

DO - 10.1109/ISBI45749.2020.9098633

M3 - Conference paper

AN - SCOPUS:85085863009

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 600

EP - 604

BT - ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

Y2 - 3 April 2020 through 7 April 2020

ER -

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454