Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework

Seyed Amir Hossein Hosseini; Burhaneddin Yaman; Chi Zhang; Kamil Ugurbil; Steen Moeller; Mehmet Akcakaya

doi:10.1109/ISBIWorkshops50223.2020.9153389

Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework

Seyed Amir Hossein Hosseini, Burhaneddin Yaman, Chi Zhang, Kamil Ugurbil, Steen Moeller, Mehmet Akcakaya

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Long scan duration remains a challenge for high-resolution MRI. Several accelerated imaging strategies have been proposed based on deep learning (DL) that require databases of fully-sampled images for training. However, scan-specific training is desired where individual variability is important, e.g. in free-breathing cardiac MRI, or where such datasets are not available due to scan time constraints for acquiring fully-sampled data. Building on our earlier method called Self-consistent Robust Artificial-neural-networks for k-space Interpolation (sRAKI), we propose a scan-specific DL reconstruction method based on recurrent neural networks that combines training and reconstruction phases of sRAKI. We use self-consistency among coils in k-space and regularization in arbitrary domains, as well as consistency with acquired data, in each iteration of the recurrent network. Results on knee MRI show that this method improves upon parallel imaging and compressed sensing methods.

Original language	English (US)
Title of host publication	ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728174013
DOIs	https://doi.org/10.1109/ISBIWorkshops50223.2020.9153389
State	Published - Apr 2020
Event	17th IEEE International Symposium on Biomedical Imaging Workshops, ISBI Workshops 2020 - Iowa City, United States Duration: Apr 4 2020 → …

Publication series

Name	ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings

Conference

Conference	17th IEEE International Symposium on Biomedical Imaging Workshops, ISBI Workshops 2020
Country/Territory	United States
City	Iowa City
Period	4/4/20 → …

Bibliographical note

Funding Information:
This work was supported by NIH U01EB025144, P41EB027061; NSF CAREER CCF-1651825. Knee MRI data were obtained from the NYU fastMRI initiative database [23]. NYU fastMRI database was acquired with the relevant institutional review board approvals as detailed in [23]. A listing of NYU fastMRI investigators, subject to updates, can be found at fastmri.med.nyu.edu.

Publisher Copyright:
© 2020 IEEE.

Keywords

Parallel imaging
compressed sensing
deep learning
image reconstruction.
machine learning
neural networks

Center for Magnetic Resonance Research (CMRR) tags

IRP

Publisher link

10.1109/ISBIWorkshops50223.2020.9153389

Find It

Find It at U of M Libraries

Cite this

Hosseini, S. A. H., Yaman, B., Zhang, C., Ugurbil, K., Moeller, S., & Akcakaya, M. (2020). Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework. In ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings Article 9153389 (ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISBIWorkshops50223.2020.9153389

Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework. / Hosseini, Seyed Amir Hossein; Yaman, Burhaneddin; Zhang, Chi et al.
ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9153389 (ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Hosseini, SAH, Yaman, B, Zhang, C, Ugurbil, K , Moeller, S & Akcakaya, M 2020, Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework. in ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings., 9153389, ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings, Institute of Electrical and Electronics Engineers Inc., 17th IEEE International Symposium on Biomedical Imaging Workshops, ISBI Workshops 2020, Iowa City, United States, 4/4/20. https://doi.org/10.1109/ISBIWorkshops50223.2020.9153389

Hosseini SAH, Yaman B, Zhang C, Ugurbil K , Moeller S , Akcakaya M. Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework. In ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. 9153389. (ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings). doi: 10.1109/ISBIWorkshops50223.2020.9153389

Hosseini, Seyed Amir Hossein ; Yaman, Burhaneddin ; Zhang, Chi et al. / Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework. ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. (ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings).

@inproceedings{7f4129bd003045c29b5ba2e4e9120638,

title = "Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework",

abstract = "Long scan duration remains a challenge for high-resolution MRI. Several accelerated imaging strategies have been proposed based on deep learning (DL) that require databases of fully-sampled images for training. However, scan-specific training is desired where individual variability is important, e.g. in free-breathing cardiac MRI, or where such datasets are not available due to scan time constraints for acquiring fully-sampled data. Building on our earlier method called Self-consistent Robust Artificial-neural-networks for k-space Interpolation (sRAKI), we propose a scan-specific DL reconstruction method based on recurrent neural networks that combines training and reconstruction phases of sRAKI. We use self-consistency among coils in k-space and regularization in arbitrary domains, as well as consistency with acquired data, in each iteration of the recurrent network. Results on knee MRI show that this method improves upon parallel imaging and compressed sensing methods.",

keywords = "Parallel imaging, compressed sensing, deep learning, image reconstruction., machine learning, neural networks",

author = "Hosseini, {Seyed Amir Hossein} and Burhaneddin Yaman and Chi Zhang and Kamil Ugurbil and Steen Moeller and Mehmet Akcakaya",

note = "Funding Information: This work was supported by NIH U01EB025144, P41EB027061; NSF CAREER CCF-1651825. Knee MRI data were obtained from the NYU fastMRI initiative database [23]. NYU fastMRI database was acquired with the relevant institutional review board approvals as detailed in [23]. A listing of NYU fastMRI investigators, subject to updates, can be found at fastmri.med.nyu.edu. Publisher Copyright: {\textcopyright} 2020 IEEE.; 17th IEEE International Symposium on Biomedical Imaging Workshops, ISBI Workshops 2020 ; Conference date: 04-04-2020",

year = "2020",

month = apr,

doi = "10.1109/ISBIWorkshops50223.2020.9153389",

language = "English (US)",

series = "ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings",

}

TY - GEN

T1 - Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework

AU - Hosseini, Seyed Amir Hossein

AU - Yaman, Burhaneddin

AU - Zhang, Chi

AU - Ugurbil, Kamil

AU - Moeller, Steen

AU - Akcakaya, Mehmet

N1 - Funding Information: This work was supported by NIH U01EB025144, P41EB027061; NSF CAREER CCF-1651825. Knee MRI data were obtained from the NYU fastMRI initiative database [23]. NYU fastMRI database was acquired with the relevant institutional review board approvals as detailed in [23]. A listing of NYU fastMRI investigators, subject to updates, can be found at fastmri.med.nyu.edu. Publisher Copyright: © 2020 IEEE.

PY - 2020/4

Y1 - 2020/4

N2 - Long scan duration remains a challenge for high-resolution MRI. Several accelerated imaging strategies have been proposed based on deep learning (DL) that require databases of fully-sampled images for training. However, scan-specific training is desired where individual variability is important, e.g. in free-breathing cardiac MRI, or where such datasets are not available due to scan time constraints for acquiring fully-sampled data. Building on our earlier method called Self-consistent Robust Artificial-neural-networks for k-space Interpolation (sRAKI), we propose a scan-specific DL reconstruction method based on recurrent neural networks that combines training and reconstruction phases of sRAKI. We use self-consistency among coils in k-space and regularization in arbitrary domains, as well as consistency with acquired data, in each iteration of the recurrent network. Results on knee MRI show that this method improves upon parallel imaging and compressed sensing methods.

AB - Long scan duration remains a challenge for high-resolution MRI. Several accelerated imaging strategies have been proposed based on deep learning (DL) that require databases of fully-sampled images for training. However, scan-specific training is desired where individual variability is important, e.g. in free-breathing cardiac MRI, or where such datasets are not available due to scan time constraints for acquiring fully-sampled data. Building on our earlier method called Self-consistent Robust Artificial-neural-networks for k-space Interpolation (sRAKI), we propose a scan-specific DL reconstruction method based on recurrent neural networks that combines training and reconstruction phases of sRAKI. We use self-consistency among coils in k-space and regularization in arbitrary domains, as well as consistency with acquired data, in each iteration of the recurrent network. Results on knee MRI show that this method improves upon parallel imaging and compressed sensing methods.

KW - Parallel imaging

KW - compressed sensing

KW - deep learning

KW - image reconstruction.

KW - machine learning

KW - neural networks

UR - http://www.scopus.com/inward/record.url?scp=85090409185&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090409185&partnerID=8YFLogxK

U2 - 10.1109/ISBIWorkshops50223.2020.9153389

DO - 10.1109/ISBIWorkshops50223.2020.9153389

M3 - Conference contribution

AN - SCOPUS:85090409185

T3 - ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings

BT - ISBI Workshops 2020 - International Symposium on Biomedical Imaging Workshops, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

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

Y2 - 4 April 2020

ER -

Scan-Specific Accelerated Mri Reconstruction Using Recurrent Neural Networks in A Regularized Self-Consistent Framework

Abstract

Publication series

Conference

Bibliographical note

Keywords

Center for Magnetic Resonance Research (CMRR) tags

Publisher link

Find It

Other files and links

Fingerprint

Cite this