Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI

David Owen; Andrew Melbourne; Zach Eaton-Rosen; David L. Thomas; Neil Marlow; Jonathan Rohrer; Sébastien Ourselin

doi:10.1007/978-3-030-00928-1_3

Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI

David Owen^*, Andrew Melbourne, Zach Eaton-Rosen, David L. Thomas, Neil Marlow, Jonathan Rohrer, Sébastien Ourselin

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

12 Citations (Scopus)

Abstract

Arterial spin labelling (ASL) is a noninvasive imaging modality, used in the clinic and in research, which can give quantitative measurements of perfusion in the brain and other organs. However, because the signal-to-noise ratio is inherently low and the ASL acquisition is particularly prone to corruption by artifact, image processing methods such as denoising and artifact filtering are vital for generating accurate measurements of perfusion. In this work, we present a new simultaneous approach to denoising and artifact removal, using a novel deep convolutional joint filter architecture to learn and exploit spatio-temporal properties of the ASL signal. We proceed to show, using data from 15 healthy subjects, that our approach achieves state of the art performance in both denoising and artifact removal, improving peak signal-to-noise ratio by up to 50%. By allowing more accurate estimation of perfusion, even in challenging datasets, this technique offers an exciting new approach for ASL pipelines, and might be used both for improving individual images and to increase the power of research studies using ASL.

Original language	English
Title of host publication	MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I
Pages	21-29
Number of pages	9
Volume	11070
DOIs	https://doi.org/10.1007/978-3-030-00928-1_3
Publication status	Published - 2018
Event	21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain Duration: 16 Sept 2018 → 20 Sept 2018

Publication series

Name	Lecture Notes in Computer Science

Conference

Conference	21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
Country/Territory	Spain
City	Granada
Period	16/09/2018 → 20/09/2018

Access to Document

10.1007/978-3-030-00928-1_3

Cite this

Owen, D., Melbourne, A., Eaton-Rosen, Z., Thomas, D. L., Marlow, N., Rohrer, J., & Ourselin, S. (2018). Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI. In MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I (Vol. 11070, pp. 21-29). (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-00928-1_3

@inbook{cdc7a748e3c74fd989a0497d47d14b75,

title = "Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI",

abstract = "Arterial spin labelling (ASL) is a noninvasive imaging modality, used in the clinic and in research, which can give quantitative measurements of perfusion in the brain and other organs. However, because the signal-to-noise ratio is inherently low and the ASL acquisition is particularly prone to corruption by artifact, image processing methods such as denoising and artifact filtering are vital for generating accurate measurements of perfusion. In this work, we present a new simultaneous approach to denoising and artifact removal, using a novel deep convolutional joint filter architecture to learn and exploit spatio-temporal properties of the ASL signal. We proceed to show, using data from 15 healthy subjects, that our approach achieves state of the art performance in both denoising and artifact removal, improving peak signal-to-noise ratio by up to 50%. By allowing more accurate estimation of perfusion, even in challenging datasets, this technique offers an exciting new approach for ASL pipelines, and might be used both for improving individual images and to increase the power of research studies using ASL.",

author = "David Owen and Andrew Melbourne and Zach Eaton-Rosen and Thomas, {David L.} and Neil Marlow and Jonathan Rohrer and S{\'e}bastien Ourselin",

year = "2018",

doi = "10.1007/978-3-030-00928-1_3",

language = "English",

isbn = "978-3-030-00927-4",

volume = "11070",

series = "Lecture Notes in Computer Science",

pages = "21--29",

booktitle = "MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I",

note = "21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 ; Conference date: 16-09-2018 Through 20-09-2018",

}

Owen, D, Melbourne, A, Eaton-Rosen, Z, Thomas, DL, Marlow, N, Rohrer, J & Ourselin, S 2018, Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI. in MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I. vol. 11070, Lecture Notes in Computer Science, pp. 21-29, 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018, Granada, Spain, 16/09/2018. https://doi.org/10.1007/978-3-030-00928-1_3

Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI. / Owen, David; Melbourne, Andrew; Eaton-Rosen, Zach et al.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I. Vol. 11070 2018. p. 21-29 (Lecture Notes in Computer Science).

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

TY - CHAP

T1 - Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI

AU - Owen, David

AU - Melbourne, Andrew

AU - Eaton-Rosen, Zach

AU - Thomas, David L.

AU - Marlow, Neil

AU - Rohrer, Jonathan

AU - Ourselin, Sébastien

PY - 2018

Y1 - 2018

N2 - Arterial spin labelling (ASL) is a noninvasive imaging modality, used in the clinic and in research, which can give quantitative measurements of perfusion in the brain and other organs. However, because the signal-to-noise ratio is inherently low and the ASL acquisition is particularly prone to corruption by artifact, image processing methods such as denoising and artifact filtering are vital for generating accurate measurements of perfusion. In this work, we present a new simultaneous approach to denoising and artifact removal, using a novel deep convolutional joint filter architecture to learn and exploit spatio-temporal properties of the ASL signal. We proceed to show, using data from 15 healthy subjects, that our approach achieves state of the art performance in both denoising and artifact removal, improving peak signal-to-noise ratio by up to 50%. By allowing more accurate estimation of perfusion, even in challenging datasets, this technique offers an exciting new approach for ASL pipelines, and might be used both for improving individual images and to increase the power of research studies using ASL.

AB - Arterial spin labelling (ASL) is a noninvasive imaging modality, used in the clinic and in research, which can give quantitative measurements of perfusion in the brain and other organs. However, because the signal-to-noise ratio is inherently low and the ASL acquisition is particularly prone to corruption by artifact, image processing methods such as denoising and artifact filtering are vital for generating accurate measurements of perfusion. In this work, we present a new simultaneous approach to denoising and artifact removal, using a novel deep convolutional joint filter architecture to learn and exploit spatio-temporal properties of the ASL signal. We proceed to show, using data from 15 healthy subjects, that our approach achieves state of the art performance in both denoising and artifact removal, improving peak signal-to-noise ratio by up to 50%. By allowing more accurate estimation of perfusion, even in challenging datasets, this technique offers an exciting new approach for ASL pipelines, and might be used both for improving individual images and to increase the power of research studies using ASL.

UR - http://www.scopus.com/inward/record.url?scp=85054091489&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00928-1_3

DO - 10.1007/978-3-030-00928-1_3

M3 - Conference paper

AN - SCOPUS:85054091489

SN - 978-3-030-00927-4

VL - 11070

T3 - Lecture Notes in Computer Science

SP - 21

EP - 29

BT - MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT I

T2 - 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018

Y2 - 16 September 2018 through 20 September 2018

ER -

Deep Convolutional Filtering for Spatio-Temporal Denoising and Artifact Removal in Arterial Spin Labelling MRI

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this