Deep learning for fast low-field MRI acquisitions

Abstract Low-field (LF) MRI research currently gains momentum from its potential to offer reduced costs and reduced footprints translating into wider accessibility. However, the impeded signal-to-noise ratio inherent to lower magnetic fields can have a significant impact on acquisition times that ch...

Full description

Bibliographic Details
Main Authors: Reina Ayde, Tobias Senft, Najat Salameh, Mathieu Sarracanie
Format: Article
Language:English
Published: Nature Portfolio 2022-07-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-022-14039-7
_version_ 1818158583303372800
author Reina Ayde
Tobias Senft
Najat Salameh
Mathieu Sarracanie
author_facet Reina Ayde
Tobias Senft
Najat Salameh
Mathieu Sarracanie
author_sort Reina Ayde
collection DOAJ
description Abstract Low-field (LF) MRI research currently gains momentum from its potential to offer reduced costs and reduced footprints translating into wider accessibility. However, the impeded signal-to-noise ratio inherent to lower magnetic fields can have a significant impact on acquisition times that challenges LF clinical relevance. Undersampling is an effective way to speed up acquisitions in MRI, and recent work has shown encouraging results when combined with deep learning (DL). Yet, training DL models generally requires large databases that are not yet available at LF regimes. Here, we demonstrate the capability of Residual U-net combined with data augmentation to reconstruct magnitude and phase information of undersampled LF MRI scans at 0.1 T with a limited training dataset (n = 10). The model performance was first evaluated in a retrospective study for different acceleration rates and sampling patterns. Ultimately, the DL approach was validated on prospectively acquired, fivefold undersampled LF data. With varying performances associated to the adopted sampling scheme, our results show that the approach investigated can preserve the global structure and the details sharpness in the reconstructed magnitude and phase images. Overall, promising results could be obtained on acquired LF MR images that may bring this research closer to clinical implementation.
first_indexed 2024-12-11T15:32:24Z
format Article
id doaj.art-7ea6e75f44c24f2a970fdd245624b61d
institution Directory Open Access Journal
issn 2045-2322
language English
last_indexed 2024-12-11T15:32:24Z
publishDate 2022-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj.art-7ea6e75f44c24f2a970fdd245624b61d2022-12-22T01:00:02ZengNature PortfolioScientific Reports2045-23222022-07-0112111310.1038/s41598-022-14039-7Deep learning for fast low-field MRI acquisitionsReina Ayde0Tobias Senft1Najat Salameh2Mathieu Sarracanie3Center for Adaptable MRI Technology (AMT Center), Department of Biomedical Engineering, University of BaselCenter for Adaptable MRI Technology (AMT Center), Department of Biomedical Engineering, University of BaselCenter for Adaptable MRI Technology (AMT Center), Department of Biomedical Engineering, University of BaselCenter for Adaptable MRI Technology (AMT Center), Department of Biomedical Engineering, University of BaselAbstract Low-field (LF) MRI research currently gains momentum from its potential to offer reduced costs and reduced footprints translating into wider accessibility. However, the impeded signal-to-noise ratio inherent to lower magnetic fields can have a significant impact on acquisition times that challenges LF clinical relevance. Undersampling is an effective way to speed up acquisitions in MRI, and recent work has shown encouraging results when combined with deep learning (DL). Yet, training DL models generally requires large databases that are not yet available at LF regimes. Here, we demonstrate the capability of Residual U-net combined with data augmentation to reconstruct magnitude and phase information of undersampled LF MRI scans at 0.1 T with a limited training dataset (n = 10). The model performance was first evaluated in a retrospective study for different acceleration rates and sampling patterns. Ultimately, the DL approach was validated on prospectively acquired, fivefold undersampled LF data. With varying performances associated to the adopted sampling scheme, our results show that the approach investigated can preserve the global structure and the details sharpness in the reconstructed magnitude and phase images. Overall, promising results could be obtained on acquired LF MR images that may bring this research closer to clinical implementation.https://doi.org/10.1038/s41598-022-14039-7
spellingShingle Reina Ayde
Tobias Senft
Najat Salameh
Mathieu Sarracanie
Deep learning for fast low-field MRI acquisitions
Scientific Reports
title Deep learning for fast low-field MRI acquisitions
title_full Deep learning for fast low-field MRI acquisitions
title_fullStr Deep learning for fast low-field MRI acquisitions
title_full_unstemmed Deep learning for fast low-field MRI acquisitions
title_short Deep learning for fast low-field MRI acquisitions
title_sort deep learning for fast low field mri acquisitions
url https://doi.org/10.1038/s41598-022-14039-7
work_keys_str_mv AT reinaayde deeplearningforfastlowfieldmriacquisitions
AT tobiassenft deeplearningforfastlowfieldmriacquisitions
AT najatsalameh deeplearningforfastlowfieldmriacquisitions
AT mathieusarracanie deeplearningforfastlowfieldmriacquisitions