CoChat
Sign Up
Cited by 313Advanced MRI Techniques and ApplicationsMedical Imaging Techniques and ApplicationsImage and Signal Denoising Methods

The DOI matches its Crossref record; 74 of the 80 references it cites were found in a registry or an index.

Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction

Open access
Muckley, Matthew J., Riemenschneider, Bruno, Radmanesh, Alireza, Kim, Sunwoo, Jeong, Geunu, Ko, Jingyu, Jun, Yohan, Shin, Hyungseob, Hwang, Dosik, Mostapha, Mahmoud, Arberet, Simon, Nickel, Dominik, Ramzi, Zaccharie, Ciuciu, Philippe, Starck, Jean-Luc, Teuwen, Jonas, Karkalousos, Dimitrios, Zhang, Chaoping, Sriram, Anuroop, Huang, Zhengnan, Yakubova, Nafissa, Lui, Yvonne W., Knoll, Florian·2021·IEEE Transactions on Medical Imaging
DOI 10.1109/tmi.2021.3075856PMC PMC8428775
Open this paper in CoChat — free accountOpen PDF

Abstract

Accelerating MRI scans is one of the principal outstanding problems in the MRI research community. Towards this goal, we hosted the second fastMRI competition targeted towards reconstructing MR images with subsampled k-space data. We provided participants with data from 7,299 clinical brain scans (de-identified via a HIPAA-compliant procedure by NYU Langone Health), holding back the fully-sampled data from 894 of these scans for challenge evaluation purposes. In contrast to the 2019 challenge, we focused our radiologist evaluations on pathological assessment in brain images. We also debuted a new Transfer track that required participants to submit models evaluated on MRI scanners from outside the training set. We received 19 submissions from eight different groups. Results showed one team scoring best in both SSIM scores and qualitative radiologist evaluations. We also performed analysis on alternative metrics to mitigate the effects of background noise and collected feedback from the participants to inform future challenges. Lastly, we identify common failure modes across the submissions, highlighting areas of need for future research in the MRI reconstruction community.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/tmi.2021.3075856

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1. Examples of 4X submissions evaluated by radiologists with slice-level SSIM scores. All methods reasonably reconstructed T2 and FLAIR images. The ATB and Neurospin methods struggled with a susceptibility region, exaggerating the focus of susceptibility and introducing a few false vessels between the susceptibility and the lateral ventricular wall. In other cases, radiologists observed mild smoothing of white matter regions on T1POST images.
Fig. 1. Examples of 4X submissions evaluated by radiologists with slice-level SSIM scores. All methods reasonably reconstructed T2 and FLAIR images. The ATB and Neurospin methods struggled with a susceptibility region, exaggerating the focus of susceptibility and introducing a few false vessels between the susceptibility and the lateral ventricular wall. In other cases, radiologists observed mild smoothing of white matter regions on T1POST images.CC BY · PubMed Central
Fig. 2. Examples of 8X submissions evaluated by radiologists with slice-level SSIM scores. At this level of acceleration fine details are smoothed and obscured for all contrasts. On T1POST images, AIRS Medical was relatively more successful than ATB and Neurospin in showing fine details of the mass, particularly in its periphery. Noticeable on the FLAIR images are horizontal “banding” effects that arise from how neural networks interact with anisotropic sampling patterns.
Fig. 2. Examples of 8X submissions evaluated by radiologists with slice-level SSIM scores. At this level of acceleration fine details are smoothed and obscured for all contrasts. On T1POST images, AIRS Medical was relatively more successful than ATB and Neurospin in showing fine details of the mass, particularly in its periphery. Noticeable on the FLAIR images are horizontal “banding” effects that arise from how neural networks interact with anisotropic sampling patterns.CC BY · PubMed Central
Fig. 3. Examples of 4X Transfer submissions evaluated by radiologists with slice-level SSIM scores. The T1POST and T2 examples are from GE scanners, whereas the FLAIR example is from a Philips scanner. All methods introduced blurring to the images. Several methods had trouble adapting to the GE data while performing relatively well on the Philips data, as seen in the form of aliasing artifacts in one of the T1POST images.
Fig. 3. Examples of 4X Transfer submissions evaluated by radiologists with slice-level SSIM scores. The T1POST and T2 examples are from GE scanners, whereas the FLAIR example is from a Philips scanner. All methods introduced blurring to the images. Several methods had trouble adapting to the GE data while performing relatively well on the Philips data, as seen in the form of aliasing artifacts in one of the T1POST images.CC BY · PubMed Central
Fig. 4. Summary of SSIM values across contestants. (top) Model perfomance for teams submitting to the main 4X and 8X Siemens competition tracks. (bottom) Model performance for teams submitting to the Transfer track (combination of GE and Philips data). The “AVG” model score for the Transfer track was a simple average across all volumes in the Transfer track.
Fig. 4. Summary of SSIM values across contestants. (top) Model perfomance for teams submitting to the main 4X and 8X Siemens competition tracks. (bottom) Model performance for teams submitting to the Transfer track (combination of GE and Philips data). The “AVG” model score for the Transfer track was a simple average across all volumes in the Transfer track.CC BY · PubMed Central
Fig. 5. Scatter plot of mean radiologist rank across cases. The horizontal axis has a separate tick for each case evaluated by the radiologist cohort. The scatter plot markers indicate whether that method was from the team with the highest, middle, or lowest SSIM scores. We generally observed radiologists awarding the best ranks to models with the best SSIM score.
Fig. 5. Scatter plot of mean radiologist rank across cases. The horizontal axis has a separate tick for each case evaluated by the radiologist cohort. The scatter plot markers indicate whether that method was from the team with the highest, middle, or lowest SSIM scores. We generally observed radiologists awarding the best ranks to models with the best SSIM score.CC BY · PubMed Central
Fig. 6. Examples of reconstruction hallucinations among challenge submissions with SSIM scores over residual plots (residuals magnified by 5). (top) A 4X submission from Neurospin generated a false vessel, possibly related to susceptibilities introduced by surgical staples. (middle) An 8X submission from ATB introduced a linear bright signal mimicking a cleft of cerebrospinal fluid, as well as blurring of the boundaries of the extra-axial mass. (bottom) A submission from ResoNNance introduced a false sulcus or prominent vessel.
Fig. 6. Examples of reconstruction hallucinations among challenge submissions with SSIM scores over residual plots (residuals magnified by 5). (top) A 4X submission from Neurospin generated a false vessel, possibly related to susceptibilities introduced by surgical staples. (middle) An 8X submission from ATB introduced a linear bright signal mimicking a cleft of cerebrospinal fluid, as well as blurring of the boundaries of the extra-axial mass. (bottom) A submission from ResoNNance introduced a false sulcus or prominent vessel.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 80 references

Research works printed but not linked: 2

  1. kotelnikov, 1933, On the carrying capacity of the ‘ether’ and wire in telecommunications, Proc 1st All-Union Conf Questions Commun
  2. uecker, 2015, Berkeley advanced reconstruction toolbox, Proc Int Soc Magn Res Med, 2486

Cited by

Sorted by citations

Items 301–312 of 312

Related

See all

Metadata

Authors
Muckley, M. J., Riemenschneider, B., Radmanesh, A., Kim, S., Jeong, G., Ko, J., Jun, Y., Shin, H., Hwang, D., Mostapha, M., Arberet, S., Nickel, D., Ramzi, Z., Ciuciu, P., Starck, J.-L., Teuwen, J., Karkalousos, D., Zhang, C., Sriram, A., Huang, Z., Yakubova, N., Lui, Y. W., & Knoll, F.
Affiliations
  • Meta (United States) (US)
  • NYU Langone Health (US)
  • Yonsei University (KR)
  • Siemens Healthcare (United States) (US)
  • Institut national de recherche en sciences et technologies du numérique (FR)
  • Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR)
  • Radboud University Nijmegen (NL)
  • Amsterdam University Medical Centers (NL)
ORCID
Venue
IEEE Transactions on Medical Imaging
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2021-09
Volume
40
Issue
9
Pages
2306-2317
ISSN
  • 0278-0062 (ISSN-L)
  • 1558-254X
DOI
10.1109/tmi.2021.3075856
PMC
PMC8428775
OpenAlex
W3157723514
Cited by
313
References
80
Citations by year
2026: 40 · 2025: 76 · 2024: 76 · 2023: 56 · 2022: 48 · 2021: 15
FWCI
25.73
Type
journal article
Language
en
Open access
Licence
Locations
  • IEEE Transactions on Medical Imaging (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Pure Amsterdam UMC (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (acceptedVersion, cc-by)
  • PubMed Central (acceptedVersion, cc-by)
  • Data Archiving and Networked Services (DANS) (submittedVersion, other-oa)
  • HAL (submittedVersion, other-oa)
  • arXiv (Cornell University)
Funders
  • NIH (R01EB024532, P41EB017183, R21EB027241)
  • NIH funding was awarded to NYU
Topics
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced MRI Techniques and Applications
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Medical Imaging Techniques and Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Signal Denoising Methods
Keywords
MRI reconstruction, accelerated MRI, brain MRI, radiologist evaluation, transfer learning, failure modes, background noise
MeSH terms
Machine Learning, Brain, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging, Neuroimaging

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review