Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge

Patricia M. Johnson; Geunu Jeong; Kerstin Hammernik; Jo Schlemper; Chen Qin; Jinming Duan; Daniel Rueckert; Jingu Lee; Nicola Pezzotti; Elwin De Weerdt; Sahar Yousefi; Mohamed S. Elmahdy; Jeroen Hendrikus Franciscus Van Gemert; Christophe Schülke; Mariya Doneva; Tim Nielsen; Sergey Kastryulin; Boudewijn P.F. Lelieveldt; Matthias J.P. Van Osch; Marius Staring; Eric Z. Chen; Puyang Wang; Xiao Chen; Terrence Chen; Vishal M. Patel; Shanhui Sun; Hyungseob Shin; Yohan Jun; Taejoon Eo; Sewon Kim; Taeseong Kim; Dosik Hwang; Patrick Putzky; Dimitrios Karkalousos; Jonas Teuwen; Nikita Miriakov; Bart Bakker; Matthan Caan; Max Welling; Matthew J. Muckley; Florian Knoll

doi:10.1007/978-3-030-88552-6_3

Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge

Patricia M. Johnson, Geunu Jeong, Kerstin Hammernik, Jo Schlemper, Chen Qin, Jinming Duan, Daniel Rueckert, Jingu Lee, Nicola Pezzotti, Elwin De Weerdt, Sahar Yousefi, Mohamed S. Elmahdy, Jeroen Hendrikus Franciscus Van Gemert, Christophe Schülke, Mariya Doneva, Tim Nielsen, Sergey Kastryulin, Boudewijn P.F. Lelieveldt, Matthias J.P. Van Osch, Marius StaringEric Z. Chen, Puyang Wang, Xiao Chen, Terrence Chen, Vishal M. Patel, Shanhui Sun, Hyungseob Shin, Yohan Jun, Taejoon Eo, Sewon Kim, Taeseong Kim, Dosik Hwang, Patrick Putzky, Dimitrios Karkalousos, Jonas Teuwen, Nikita Miriakov, Bart Bakker, Matthan Caan, Max Welling, Matthew J. Muckley, Florian Knoll

Department of Electrical and Electronic Engineering

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

1 Citation (Scopus)

Abstract

The 2019 fastMRI challenge was an open challenge designed to advance research in the field of machine learning for MR image reconstruction. The goal for the participants was to reconstruct undersampled MRI k-space data. The original challenge left an open question as to how well the reconstruction methods will perform in the setting where there is a systematic difference between training and test data. In this work we tested the generalization performance of the submissions with respect to various perturbations, and despite differences in model architecture and training, all of the methods perform very similarly.

Original language	English
Title of host publication	Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings
Editors	Nandinee Haq, Patricia Johnson, Andreas Maier, Tobias Würfl, Jaejun Yoo
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	25-34
Number of pages	10
ISBN (Print)	9783030885519
DOIs	https://doi.org/10.1007/978-3-030-88552-6_3
Publication status	Published - 2021
Event	4th International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2021 held in Conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online Duration: 2021 Oct 1 → 2021 Oct 1

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12964 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2021 held in Conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
City	Virtual, Online
Period	21/10/1 → 21/10/1

Bibliographical note

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-030-88552-6_3

Cite this

Johnson, P. M., Jeong, G., Hammernik, K., Schlemper, J., Qin, C., Duan, J., Rueckert, D., Lee, J., Pezzotti, N., De Weerdt, E., Yousefi, S., Elmahdy, M. S., Van Gemert, J. H. F., Schülke, C., Doneva, M., Nielsen, T., Kastryulin, S., Lelieveldt, B. P. F., Van Osch, M. J. P., ... Knoll, F. (2021). Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. In N. Haq, P. Johnson, A. Maier, T. Würfl, & J. Yoo (Eds.), Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings (pp. 25-34). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12964 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-88552-6_3

Johnson, Patricia M. ; Jeong, Geunu ; Hammernik, Kerstin et al. / Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings. editor / Nandinee Haq ; Patricia Johnson ; Andreas Maier ; Tobias Würfl ; Jaejun Yoo. Springer Science and Business Media Deutschland GmbH, 2021. pp. 25-34 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{3d08e0b64783465db619d2308ee93093,

title = "Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge",

abstract = "The 2019 fastMRI challenge was an open challenge designed to advance research in the field of machine learning for MR image reconstruction. The goal for the participants was to reconstruct undersampled MRI k-space data. The original challenge left an open question as to how well the reconstruction methods will perform in the setting where there is a systematic difference between training and test data. In this work we tested the generalization performance of the submissions with respect to various perturbations, and despite differences in model architecture and training, all of the methods perform very similarly.",

author = "Johnson, {Patricia M.} and Geunu Jeong and Kerstin Hammernik and Jo Schlemper and Chen Qin and Jinming Duan and Daniel Rueckert and Jingu Lee and Nicola Pezzotti and {De Weerdt}, Elwin and Sahar Yousefi and Elmahdy, {Mohamed S.} and {Van Gemert}, {Jeroen Hendrikus Franciscus} and Christophe Sch{\"u}lke and Mariya Doneva and Tim Nielsen and Sergey Kastryulin and Lelieveldt, {Boudewijn P.F.} and {Van Osch}, {Matthias J.P.} and Marius Staring and Chen, {Eric Z.} and Puyang Wang and Xiao Chen and Terrence Chen and Patel, {Vishal M.} and Shanhui Sun and Hyungseob Shin and Yohan Jun and Taejoon Eo and Sewon Kim and Taeseong Kim and Dosik Hwang and Patrick Putzky and Dimitrios Karkalousos and Jonas Teuwen and Nikita Miriakov and Bart Bakker and Matthan Caan and Max Welling and Muckley, {Matthew J.} and Florian Knoll",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 4th International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2021 held in Conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 ; Conference date: 01-10-2021 Through 01-10-2021",

year = "2021",

doi = "10.1007/978-3-030-88552-6_3",

language = "English",

isbn = "9783030885519",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "25--34",

editor = "Nandinee Haq and Patricia Johnson and Andreas Maier and Tobias W{\"u}rfl and Jaejun Yoo",

booktitle = "Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings",

address = "Germany",

}

Johnson, PM, Jeong, G, Hammernik, K, Schlemper, J, Qin, C, Duan, J, Rueckert, D, Lee, J, Pezzotti, N, De Weerdt, E, Yousefi, S, Elmahdy, MS, Van Gemert, JHF, Schülke, C, Doneva, M, Nielsen, T, Kastryulin, S, Lelieveldt, BPF, Van Osch, MJP, Staring, M, Chen, EZ, Wang, P, Chen, X, Chen, T, Patel, VM, Sun, S, Shin, H, Jun, Y, Eo, T, Kim, S, Kim, T, Hwang, D, Putzky, P, Karkalousos, D, Teuwen, J, Miriakov, N, Bakker, B, Caan, M, Welling, M, Muckley, MJ & Knoll, F 2021, Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. in N Haq, P Johnson, A Maier, T Würfl & J Yoo (eds), Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12964 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 25-34, 4th International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2021 held in Conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, Virtual, Online, 21/10/1. https://doi.org/10.1007/978-3-030-88552-6_3

Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. / Johnson, Patricia M.; Jeong, Geunu; Hammernik, Kerstin et al.

Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings. ed. / Nandinee Haq; Patricia Johnson; Andreas Maier; Tobias Würfl; Jaejun Yoo. Springer Science and Business Media Deutschland GmbH, 2021. p. 25-34 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12964 LNCS).

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

TY - GEN

T1 - Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge

AU - Johnson, Patricia M.

AU - Jeong, Geunu

AU - Hammernik, Kerstin

AU - Schlemper, Jo

AU - Qin, Chen

AU - Duan, Jinming

AU - Rueckert, Daniel

AU - Lee, Jingu

AU - Pezzotti, Nicola

AU - De Weerdt, Elwin

AU - Yousefi, Sahar

AU - Elmahdy, Mohamed S.

AU - Van Gemert, Jeroen Hendrikus Franciscus

AU - Schülke, Christophe

AU - Doneva, Mariya

AU - Nielsen, Tim

AU - Kastryulin, Sergey

AU - Lelieveldt, Boudewijn P.F.

AU - Van Osch, Matthias J.P.

AU - Staring, Marius

AU - Chen, Eric Z.

AU - Wang, Puyang

AU - Chen, Xiao

AU - Chen, Terrence

AU - Patel, Vishal M.

AU - Sun, Shanhui

AU - Shin, Hyungseob

AU - Jun, Yohan

AU - Eo, Taejoon

AU - Kim, Sewon

AU - Kim, Taeseong

AU - Hwang, Dosik

AU - Putzky, Patrick

AU - Karkalousos, Dimitrios

AU - Teuwen, Jonas

AU - Miriakov, Nikita

AU - Bakker, Bart

AU - Caan, Matthan

AU - Welling, Max

AU - Muckley, Matthew J.

AU - Knoll, Florian

PY - 2021

Y1 - 2021

N2 - The 2019 fastMRI challenge was an open challenge designed to advance research in the field of machine learning for MR image reconstruction. The goal for the participants was to reconstruct undersampled MRI k-space data. The original challenge left an open question as to how well the reconstruction methods will perform in the setting where there is a systematic difference between training and test data. In this work we tested the generalization performance of the submissions with respect to various perturbations, and despite differences in model architecture and training, all of the methods perform very similarly.

AB - The 2019 fastMRI challenge was an open challenge designed to advance research in the field of machine learning for MR image reconstruction. The goal for the participants was to reconstruct undersampled MRI k-space data. The original challenge left an open question as to how well the reconstruction methods will perform in the setting where there is a systematic difference between training and test data. In this work we tested the generalization performance of the submissions with respect to various perturbations, and despite differences in model architecture and training, all of the methods perform very similarly.

UR - http://www.scopus.com/inward/record.url?scp=85116879187&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85116879187&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-88552-6_3

DO - 10.1007/978-3-030-88552-6_3

M3 - Conference contribution

AN - SCOPUS:85116879187

SN - 9783030885519

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 25

EP - 34

BT - Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings

A2 - Haq, Nandinee

A2 - Johnson, Patricia

A2 - Maier, Andreas

A2 - Würfl, Tobias

A2 - Yoo, Jaejun

PB - Springer Science and Business Media Deutschland GmbH

T2 - 4th International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2021 held in Conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021

Y2 - 1 October 2021 through 1 October 2021

ER -

Johnson PM, Jeong G, Hammernik K, Schlemper J, Qin C, Duan J et al. Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. In Haq N, Johnson P, Maier A, Würfl T, Yoo J, editors, Machine Learning for Medical Image Reconstruction - 4th International Workshop, MLMIR 2021, Held in Conjunction with MICCAI 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 25-34. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-88552-6_3

Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this