Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI

Patricia M. Johnson; Dana J. Lin; Jure Zbontar; C. Lawrence Zitnick; Anuroop Sriram; Matthew Muckley; James S. Babb; Mitchell Kline; Gina Ciavarra; Erin Alaia; Mohammad Samim; William R. Walter; Liz Calderon; Thomas Pock; Daniel K. Sodickson; Michael P. Recht; Florian Knoll

doi:10.1148/radiol.220425

Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI

Patricia M. Johnson, Dana J. Lin, Jure Zbontar, C. Lawrence Zitnick, Anuroop Sriram, Matthew Muckley, James S. Babb, Mitchell Kline, Gina Ciavarra, Erin Alaia, Mohammad Samim, William R. Walter, Liz Calderon, Thomas Pock, Daniel K. Sodickson, Michael P. Recht, Florian Knoll

Biomedical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Background: MRI is a powerful diagnostic tool with a long acquisition time. Recently, deep learning (DL) methods have provided accelerated high-quality image reconstructions from undersampled data, but it is unclear if DL image reconstruction can be reliably translated to everyday clinical practice. Purpose: To determine the diagnostic equivalence of prospectively accelerated DL-reconstructed knee MRI compared with conventional accelerated MRI for evaluating internal derangement of the knee in a clinical setting. Materials and Methods: A DL reconstruction model was trained with images from 298 clinical 3-T knee examinations. In a prospective analysis, patients clinically referred for knee MRI underwent a conventional accelerated knee MRI protocol at 3 T followed by an accelerated DL protocol between January 2020 and February 2021. The equivalence of the DL reconstruction of the images relative to the conventional images for the detection of an abnormality was assessed in terms of interchangeability. Each examination was reviewed by six musculoskeletal radiologists. Analyses pertaining to the detection of meniscal or ligament tears and bone marrow or cartilage abnormalities were based on four-point ordinal scores for the likelihood of an abnormality. Additionally, the protocols were compared with use of four-point ordinal scores for each aspect of image quality: overall image quality, presence of artifacts, sharpness, and signal-to-noise ratio. Results: A total of 170 participants (mean age ± SD, 45 years ± 16; 76 men) were evaluated. The DL-reconstructed images were determined to be of diagnostic equivalence with the conventional images for detection of abnormalities. The overall image quality score, averaged over six readers, was significantly better (P < .001) for the DL than for the conventional images. Conclusion: In a clinical setting, deep learning reconstruction enabled a nearly twofold reduction in scan time for a knee MRI and was diagnostically equivalent with the conventional protocol.

Original language	English (US)
Article number	e220425
Journal	Radiology
Volume	307
Issue number	2
DOIs	https://doi.org/10.1148/radiol.220425
State	Published - Apr 2023

ASJC Scopus subject areas

Radiology Nuclear Medicine and imaging

Access to Document

10.1148/radiol.220425

Cite this

Johnson, P. M., Lin, D. J., Zbontar, J., Zitnick, C. L., Sriram, A., Muckley, M., Babb, J. S., Kline, M., Ciavarra, G., Alaia, E., Samim, M., Walter, W. R., Calderon, L., Pock, T., Sodickson, D. K., Recht, M. P., & Knoll, F. (2023). Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI. Radiology, 307(2), Article e220425. https://doi.org/10.1148/radiol.220425

@article{539d7bf27aa149e69453359b3989f31a,

title = "Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI",

abstract = "Background: MRI is a powerful diagnostic tool with a long acquisition time. Recently, deep learning (DL) methods have provided accelerated high-quality image reconstructions from undersampled data, but it is unclear if DL image reconstruction can be reliably translated to everyday clinical practice. Purpose: To determine the diagnostic equivalence of prospectively accelerated DL-reconstructed knee MRI compared with conventional accelerated MRI for evaluating internal derangement of the knee in a clinical setting. Materials and Methods: A DL reconstruction model was trained with images from 298 clinical 3-T knee examinations. In a prospective analysis, patients clinically referred for knee MRI underwent a conventional accelerated knee MRI protocol at 3 T followed by an accelerated DL protocol between January 2020 and February 2021. The equivalence of the DL reconstruction of the images relative to the conventional images for the detection of an abnormality was assessed in terms of interchangeability. Each examination was reviewed by six musculoskeletal radiologists. Analyses pertaining to the detection of meniscal or ligament tears and bone marrow or cartilage abnormalities were based on four-point ordinal scores for the likelihood of an abnormality. Additionally, the protocols were compared with use of four-point ordinal scores for each aspect of image quality: overall image quality, presence of artifacts, sharpness, and signal-to-noise ratio. Results: A total of 170 participants (mean age ± SD, 45 years ± 16; 76 men) were evaluated. The DL-reconstructed images were determined to be of diagnostic equivalence with the conventional images for detection of abnormalities. The overall image quality score, averaged over six readers, was significantly better (P < .001) for the DL than for the conventional images. Conclusion: In a clinical setting, deep learning reconstruction enabled a nearly twofold reduction in scan time for a knee MRI and was diagnostically equivalent with the conventional protocol.",

author = "Johnson, {Patricia M.} and Lin, {Dana J.} and Jure Zbontar and Zitnick, {C. Lawrence} and Anuroop Sriram and Matthew Muckley and Babb, {James S.} and Mitchell Kline and Gina Ciavarra and Erin Alaia and Mohammad Samim and Walter, {William R.} and Liz Calderon and Thomas Pock and Sodickson, {Daniel K.} and Recht, {Michael P.} and Florian Knoll",

note = "Publisher Copyright: {\textcopyright} RSNA, 2023.",

year = "2023",

month = apr,

doi = "10.1148/radiol.220425",

language = "English (US)",

volume = "307",

journal = "Radiology",

issn = "0033-8419",

publisher = "Radiological Society of North America Inc.",

number = "2",

}

TY - JOUR

T1 - Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI

AU - Johnson, Patricia M.

AU - Lin, Dana J.

AU - Zbontar, Jure

AU - Zitnick, C. Lawrence

AU - Sriram, Anuroop

AU - Muckley, Matthew

AU - Babb, James S.

AU - Kline, Mitchell

AU - Ciavarra, Gina

AU - Alaia, Erin

AU - Samim, Mohammad

AU - Walter, William R.

AU - Calderon, Liz

AU - Pock, Thomas

AU - Sodickson, Daniel K.

AU - Recht, Michael P.

AU - Knoll, Florian

PY - 2023/4

Y1 - 2023/4

N2 - Background: MRI is a powerful diagnostic tool with a long acquisition time. Recently, deep learning (DL) methods have provided accelerated high-quality image reconstructions from undersampled data, but it is unclear if DL image reconstruction can be reliably translated to everyday clinical practice. Purpose: To determine the diagnostic equivalence of prospectively accelerated DL-reconstructed knee MRI compared with conventional accelerated MRI for evaluating internal derangement of the knee in a clinical setting. Materials and Methods: A DL reconstruction model was trained with images from 298 clinical 3-T knee examinations. In a prospective analysis, patients clinically referred for knee MRI underwent a conventional accelerated knee MRI protocol at 3 T followed by an accelerated DL protocol between January 2020 and February 2021. The equivalence of the DL reconstruction of the images relative to the conventional images for the detection of an abnormality was assessed in terms of interchangeability. Each examination was reviewed by six musculoskeletal radiologists. Analyses pertaining to the detection of meniscal or ligament tears and bone marrow or cartilage abnormalities were based on four-point ordinal scores for the likelihood of an abnormality. Additionally, the protocols were compared with use of four-point ordinal scores for each aspect of image quality: overall image quality, presence of artifacts, sharpness, and signal-to-noise ratio. Results: A total of 170 participants (mean age ± SD, 45 years ± 16; 76 men) were evaluated. The DL-reconstructed images were determined to be of diagnostic equivalence with the conventional images for detection of abnormalities. The overall image quality score, averaged over six readers, was significantly better (P < .001) for the DL than for the conventional images. Conclusion: In a clinical setting, deep learning reconstruction enabled a nearly twofold reduction in scan time for a knee MRI and was diagnostically equivalent with the conventional protocol.

AB - Background: MRI is a powerful diagnostic tool with a long acquisition time. Recently, deep learning (DL) methods have provided accelerated high-quality image reconstructions from undersampled data, but it is unclear if DL image reconstruction can be reliably translated to everyday clinical practice. Purpose: To determine the diagnostic equivalence of prospectively accelerated DL-reconstructed knee MRI compared with conventional accelerated MRI for evaluating internal derangement of the knee in a clinical setting. Materials and Methods: A DL reconstruction model was trained with images from 298 clinical 3-T knee examinations. In a prospective analysis, patients clinically referred for knee MRI underwent a conventional accelerated knee MRI protocol at 3 T followed by an accelerated DL protocol between January 2020 and February 2021. The equivalence of the DL reconstruction of the images relative to the conventional images for the detection of an abnormality was assessed in terms of interchangeability. Each examination was reviewed by six musculoskeletal radiologists. Analyses pertaining to the detection of meniscal or ligament tears and bone marrow or cartilage abnormalities were based on four-point ordinal scores for the likelihood of an abnormality. Additionally, the protocols were compared with use of four-point ordinal scores for each aspect of image quality: overall image quality, presence of artifacts, sharpness, and signal-to-noise ratio. Results: A total of 170 participants (mean age ± SD, 45 years ± 16; 76 men) were evaluated. The DL-reconstructed images were determined to be of diagnostic equivalence with the conventional images for detection of abnormalities. The overall image quality score, averaged over six readers, was significantly better (P < .001) for the DL than for the conventional images. Conclusion: In a clinical setting, deep learning reconstruction enabled a nearly twofold reduction in scan time for a knee MRI and was diagnostically equivalent with the conventional protocol.

UR - http://www.scopus.com/inward/record.url?scp=85152173566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85152173566&partnerID=8YFLogxK

U2 - 10.1148/radiol.220425

DO - 10.1148/radiol.220425

M3 - Article

C2 - 36648347

AN - SCOPUS:85152173566

SN - 0033-8419

VL - 307

JO - Radiology

JF - Radiology

IS - 2

M1 - e220425

ER -

Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this