3D BBPConvNet to reconstruct parallel MRI

Kyong Hwan Jin; Michael Unser

doi:10.1109/ISBI.2018.8363593

3D BBPConvNet to reconstruct parallel MRI

Kyong Hwan Jin, Michael Unser

Department of Electrical Engineering and Computer Science

Swiss Federal Institute of Technology Lausanne

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

8 Scopus citations

Abstract

In recent years, compressed sensing techniques have been applied to the reconstruction of parallel magnetic resonance (MR) images. Particularly for 3D MR signal, it is crucial to acquire fewer samples to reduce the distortions caused by long-time acquisitions (e.g., motion, organ dynamics). Motivated by the recent success of ConvNet in 2D image reconstruction, we propose to extend the approach to 3D volume reconstruction and parallel MR imaging. The structure of the proposed network follows FBPConvNet with additional coil compression by SSoS and wavelet transform. A parallelism using two GPUs is also applied to overcome the memory shortage. The proposed method is able to reconstruct a (320 × 320 × 256 × 8) volume in less than 10s with 2 GPUs, while the iterative algorithm ℓ₁-ESPIRiT takes over 5 min in CPU.

Original language	English
Title of host publication	2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
Publisher	IEEE Computer Society
Pages	361-364
Number of pages	4
ISBN (Electronic)	9781538636367
DOIs	https://doi.org/10.1109/ISBI.2018.8363593
State	Published - 23 May 2018
Event	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States Duration: 4 Apr 2018 → 7 Apr 2018

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2018-April
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
Country/Territory	United States
City	Washington
Period	4/04/18 → 7/04/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

3D MRI reconstruction
Convolutional neural network
FBPConvNet
Parallel MRI

Access to Document

10.1109/ISBI.2018.8363593

Cite this

@inproceedings{2fcc5d6a9e304bb4899c091fef861cb3,

title = "3D BBPConvNet to reconstruct parallel MRI",

abstract = "In recent years, compressed sensing techniques have been applied to the reconstruction of parallel magnetic resonance (MR) images. Particularly for 3D MR signal, it is crucial to acquire fewer samples to reduce the distortions caused by long-time acquisitions (e.g., motion, organ dynamics). Motivated by the recent success of ConvNet in 2D image reconstruction, we propose to extend the approach to 3D volume reconstruction and parallel MR imaging. The structure of the proposed network follows FBPConvNet with additional coil compression by SSoS and wavelet transform. A parallelism using two GPUs is also applied to overcome the memory shortage. The proposed method is able to reconstruct a (320 × 320 × 256 × 8) volume in less than 10s with 2 GPUs, while the iterative algorithm ℓ1-ESPIRiT takes over 5 min in CPU.",

keywords = "3D MRI reconstruction, Convolutional neural network, FBPConvNet, Parallel MRI",

author = "Jin, {Kyong Hwan} and Michael Unser",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 ; Conference date: 04-04-2018 Through 07-04-2018",

year = "2018",

month = may,

day = "23",

doi = "10.1109/ISBI.2018.8363593",

language = "English",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

pages = "361--364",

booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018",

}

Jin, KH & Unser, M 2018, 3D BBPConvNet to reconstruct parallel MRI. in 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Proceedings - International Symposium on Biomedical Imaging, vol. 2018-April, IEEE Computer Society, pp. 361-364, 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018, Washington, United States, 4/04/18. https://doi.org/10.1109/ISBI.2018.8363593

3D BBPConvNet to reconstruct parallel MRI. / Jin, Kyong Hwan; Unser, Michael.
2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. IEEE Computer Society, 2018. p. 361-364 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2018-April).

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

TY - GEN

T1 - 3D BBPConvNet to reconstruct parallel MRI

AU - Jin, Kyong Hwan

AU - Unser, Michael

PY - 2018/5/23

Y1 - 2018/5/23

N2 - In recent years, compressed sensing techniques have been applied to the reconstruction of parallel magnetic resonance (MR) images. Particularly for 3D MR signal, it is crucial to acquire fewer samples to reduce the distortions caused by long-time acquisitions (e.g., motion, organ dynamics). Motivated by the recent success of ConvNet in 2D image reconstruction, we propose to extend the approach to 3D volume reconstruction and parallel MR imaging. The structure of the proposed network follows FBPConvNet with additional coil compression by SSoS and wavelet transform. A parallelism using two GPUs is also applied to overcome the memory shortage. The proposed method is able to reconstruct a (320 × 320 × 256 × 8) volume in less than 10s with 2 GPUs, while the iterative algorithm ℓ1-ESPIRiT takes over 5 min in CPU.

AB - In recent years, compressed sensing techniques have been applied to the reconstruction of parallel magnetic resonance (MR) images. Particularly for 3D MR signal, it is crucial to acquire fewer samples to reduce the distortions caused by long-time acquisitions (e.g., motion, organ dynamics). Motivated by the recent success of ConvNet in 2D image reconstruction, we propose to extend the approach to 3D volume reconstruction and parallel MR imaging. The structure of the proposed network follows FBPConvNet with additional coil compression by SSoS and wavelet transform. A parallelism using two GPUs is also applied to overcome the memory shortage. The proposed method is able to reconstruct a (320 × 320 × 256 × 8) volume in less than 10s with 2 GPUs, while the iterative algorithm ℓ1-ESPIRiT takes over 5 min in CPU.

KW - 3D MRI reconstruction

KW - Convolutional neural network

KW - FBPConvNet

KW - Parallel MRI

UR - http://www.scopus.com/inward/record.url?scp=85048136611&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2018.8363593

DO - 10.1109/ISBI.2018.8363593

M3 - Conference contribution

AN - SCOPUS:85048136611

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 361

EP - 364

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018

PB - IEEE Computer Society

T2 - 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018

Y2 - 4 April 2018 through 7 April 2018

ER -

3D BBPConvNet to reconstruct parallel MRI

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this