Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach
Echo planar imaging (EPI), a fast magnetic resonance imaging technique, is a powerful tool in functional neuroimaging studies. However, susceptibility artifacts, which cause misinterpretations of brain functions, are unavoidable distortions in EPI. This paper proposes an end-to-end deep learning fra...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/7/2314 |
id |
doaj-a8b0d57a5f944d80a76680de71c53c77 |
---|---|
record_format |
Article |
spelling |
doaj-a8b0d57a5f944d80a76680de71c53c772021-03-27T00:01:45ZengMDPI AGSensors1424-82202021-03-01212314231410.3390/s21072314Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning ApproachSoan T. M. Duong0Son Lam Phung1Abdesselam Bouzerdoum2Sui Paul Ang3Mark M. Schira4School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, Wollongong, NSW 2522, AustraliaSchool of Electrical, Computer and Telecommunications Engineering, University of Wollongong, Wollongong, NSW 2522, AustraliaSchool of Electrical, Computer and Telecommunications Engineering, University of Wollongong, Wollongong, NSW 2522, AustraliaSchool of Electrical, Computer and Telecommunications Engineering, University of Wollongong, Wollongong, NSW 2522, AustraliaSchool of Psychology, University of Wollongong, Wollongong, NSW 2522, AustraliaEcho planar imaging (EPI), a fast magnetic resonance imaging technique, is a powerful tool in functional neuroimaging studies. However, susceptibility artifacts, which cause misinterpretations of brain functions, are unavoidable distortions in EPI. This paper proposes an end-to-end deep learning framework, named TS-Net, for susceptibility artifact correction (SAC) in a pair of 3D EPI images with reversed phase-encoding directions. The proposed TS-Net comprises a deep convolutional network to predict a displacement field in three dimensions to overcome the limitation of existing methods, which only estimate the displacement field along the dominant-distortion direction. In the training phase, anatomical T1-weighted images are leveraged to regularize the correction, but they are not required during the inference phase to make TS-Net more flexible for general use. The experimental results show that TS-Net achieves favorable accuracy and speed trade-off when compared with the state-of-the-art SAC methods, i.e., TOPUP, TISAC, and S-Net. The fast inference speed (less than a second) of TS-Net makes real-time SAC during EPI image acquisition feasible and accelerates the medical image-processing pipelines.https://www.mdpi.com/1424-8220/21/7/2314susceptibility artifactsdeep learninghigh-speedecho planar imagingreversed phase-encoding |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Soan T. M. Duong Son Lam Phung Abdesselam Bouzerdoum Sui Paul Ang Mark M. Schira |
spellingShingle |
Soan T. M. Duong Son Lam Phung Abdesselam Bouzerdoum Sui Paul Ang Mark M. Schira Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach Sensors susceptibility artifacts deep learning high-speed echo planar imaging reversed phase-encoding |
author_facet |
Soan T. M. Duong Son Lam Phung Abdesselam Bouzerdoum Sui Paul Ang Mark M. Schira |
author_sort |
Soan T. M. Duong |
title |
Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach |
title_short |
Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach |
title_full |
Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach |
title_fullStr |
Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach |
title_full_unstemmed |
Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach |
title_sort |
correcting susceptibility artifacts of mri sensors in brain scanning: a 3d anatomy-guided deep learning approach |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2021-03-01 |
description |
Echo planar imaging (EPI), a fast magnetic resonance imaging technique, is a powerful tool in functional neuroimaging studies. However, susceptibility artifacts, which cause misinterpretations of brain functions, are unavoidable distortions in EPI. This paper proposes an end-to-end deep learning framework, named TS-Net, for susceptibility artifact correction (SAC) in a pair of 3D EPI images with reversed phase-encoding directions. The proposed TS-Net comprises a deep convolutional network to predict a displacement field in three dimensions to overcome the limitation of existing methods, which only estimate the displacement field along the dominant-distortion direction. In the training phase, anatomical T1-weighted images are leveraged to regularize the correction, but they are not required during the inference phase to make TS-Net more flexible for general use. The experimental results show that TS-Net achieves favorable accuracy and speed trade-off when compared with the state-of-the-art SAC methods, i.e., TOPUP, TISAC, and S-Net. The fast inference speed (less than a second) of TS-Net makes real-time SAC during EPI image acquisition feasible and accelerates the medical image-processing pipelines. |
topic |
susceptibility artifacts deep learning high-speed echo planar imaging reversed phase-encoding |
url |
https://www.mdpi.com/1424-8220/21/7/2314 |
work_keys_str_mv |
AT soantmduong correctingsusceptibilityartifactsofmrisensorsinbrainscanninga3danatomyguideddeeplearningapproach AT sonlamphung correctingsusceptibilityartifactsofmrisensorsinbrainscanninga3danatomyguideddeeplearningapproach AT abdesselambouzerdoum correctingsusceptibilityartifactsofmrisensorsinbrainscanninga3danatomyguideddeeplearningapproach AT suipaulang correctingsusceptibilityartifactsofmrisensorsinbrainscanninga3danatomyguideddeeplearningapproach AT markmschira correctingsusceptibilityartifactsofmrisensorsinbrainscanninga3danatomyguideddeeplearningapproach |
_version_ |
1724201740633374720 |