Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Cited by 201

The DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.

SynthMorph: Learning Contrast-Invariant Registration Without Acquired Images

Open access
Hoffmann, Malte, Billot, Benjamin, Greve, Douglas N., Iglesias, Juan Eugenio, Fischl, Bruce, Dalca, Adrian V. · 2022 · IEEE Transactions on Medical Imaging

Abstract

We introduce a strategy for learning image registration without acquired imaging data, producing powerful networks agnostic to contrast introduced by magnetic resonance imaging (MRI). While classical registration methods accurately estimate the spatial correspondence between images, they solve an optimization problem for every new image pair. Learning-based techniques are fast at test time but limited to registering images with contrasts and geometric content similar to those seen during training. We propose to remove this dependency on training data by leveraging a generative strategy for diverse synthetic label maps and images that exposes networks to a wide range of variability, forcing them to learn more invariant features. This approach results in powerful networks that accurately generalize to a broad array of MRI contrasts. We present extensive experiments with a focus on 3D neuroimaging, showing that this strategy enables robust and accurate registration of arbitrary MRI contrasts even if the target contrast is not seen by the networks during training. We demonstrate registration accuracy surpassing the state of the art both within and across contrasts, using a single model. Critically, training on arbitrary shapes synthesized from noise distributions results in competitive performance, removing the dependency on acquired data of any kind. Additionally, since anatomical label maps are often available for the anatomy of interest, we show that synthesizing images from these dramatically boosts performance, while still avoiding the need for real intensity images. Our code is available at doic https://w3id.org/synthmorph.

Overview

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

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

Integrity signals

Registry

Crossref · 10.1109/tmi.2021.3116879

Cites

14 of 14 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

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

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

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 14 references

Research works printed but not linked: 84

  1. rühaak, 2013, A fully parallel algorithm for multimodal image registration using normalized gradient fields, Proc ISBI, 572
  2. 10.1109/ISBI.2015.7163977 · DOI
  3. 10.1016/j.media.2012.05.008 · DOI · Paper page
  4. 10.1007/11866763_89 · DOI · Paper page
  5. van ginneken, 2019, Medical Image Computing and Computer-Assisted Intervention Curious
  6. 10.1109/TMI.2019.2935060 · DOI
  7. könig, 2014, A fast and accurate parallel algorithm for non-linear image registration using normalized gradient fields, Proc ISBI, 580
  8. 10.1109/TBME.2016.2574816 · DOI · Paper page
  9. 10.1016/j.media.2011.03.001 · DOI · Paper page
  10. 10.1016/j.media.2005.01.002 · DOI
  11. 10.1007/BFb0056301 · DOI · Paper page
  12. 10.1023/A:1007958904918 · DOI · Paper page
  13. 10.1007/978-3-642-40811-3_79 · DOI
  14. li, 2017, Non-rigid image registration using fully convolutional networks with deep self-supervision, arXiv 1709 00799
  15. 10.1007/978-3-319-66182-7_27 · DOI · Paper page
  16. 10.1109/TBME.2015.2496253 · DOI · Paper page
  17. 10.1023/B:VISI.0000043755.93987.aa · DOI · Paper page
  18. roy, 2014, MR to CT registration of brains using image synthesis, Proc SPIE Med Ima Image Process, 9034
  19. 10.1016/j.neuroimage.2015.03.050 · DOI
  20. 10.1007/978-3-030-20351-1_3 · DOI
  21. lee, 2019, Few labeled atlases are necessary for deep-learning-based segmentation, Proc Mach Learn Health, 1
  22. 10.1016/j.neunet.2020.01.023 · DOI
  23. 10.1007/978-3-658-25326-4_69 · DOI
  24. 10.1109/ISBI.2018.8363756 · DOI
  25. 10.1016/j.media.2018.07.002 · DOI · Paper page
  26. 10.1007/978-3-030-20351-1_19 · DOI
  27. tanner, 2018, Generative adversarial networks for MR-CT deformable image registration, arXiv 1807 07349
  28. 10.1016/j.patcog.2016.09.026 · DOI
  29. 10.1201/9781420042474 · doi:10.1201/9781420042474
  30. 10.1016/j.neuroimage.2016.09.011 · DOI · Paper page
  31. frackowiak, 2004, Human Brain Function
  32. 10.1038/s41467-019-13374-0 · DOI
  33. 10.1016/j.media.2016.10.005 · DOI
  34. 10.1007/978-3-319-02126-3_3 · DOI
  35. 10.1002/cne.24946 · DOI
  36. 10.1109/TMI.2019.2897112 · DOI · Paper page
  37. 10.1109/TMI.2018.2878316 · DOI · Paper page
  38. 10.1016/j.media.2015.09.005 · DOI
  39. 10.1109/ISBI.2017.7950652 · DOI
  40. 10.1109/TMI.2002.806587 · DOI · Paper page
  41. 10.1109/42.906424 · DOI · Paper page
  42. 10.1109/42.511747 · DOI · Paper page
  43. 10.1109/42.811270 · DOI · Paper page
  44. abadi, 2016, Tensorflow: A system for large-scale machine learning, Proc USENIX Symp on Operating System Design and Implementation, 265
  45. kingma, 2014, Adam: A method for stochastic optimization, arXiv 1412 6980
  46. 10.1016/j.media.2005.09.004 · DOI · Paper page
  47. 10.1155/IJBI/2006/49515 · DOI · Paper page
  48. 10.1162/jocn.2007.19.9.1498 · DOI · Paper page
  49. 10.1016/j.neuroimage.2018.10.009 · DOI · Paper page
  50. zhao, 2019, Data augmentation using learned transforms for one-shot medical image segmentation, CoRR, abs 1902 9383, 1
  51. 10.1007/978-3-030-32251-9_44 · doi:10.1007/978-3-030-32251-9_44
  52. 10.1007/978-3-319-59050-9_47 · DOI · Paper page
  53. billot, 2020, A learning strategy for contrast-agnostic mri segmentation, Proc PMLR, 121, 75
  54. 10.1109/ISBI48211.2021.9434113 · DOI
  55. dalca, 2018, VoxelMorph Learning-Based Image Registration
  56. 10.1007/11866565_113 · DOI · Paper page
  57. 10.1016/j.mri.2014.11.004 · doi:10.1016/j.mri.2014.11.004
  58. 10.1017/9781107706958 · DOI · Paper page
  59. shi, 2012, A comprehensive cardiac motion estimation framework using both untagged and 3-D tagged MR images based on nonrigid registration, IEEE Trans Med Imag, 31, 1263, 10.1109/TMI.2012.2188104 · DOI
  60. 10.2307/1932409 · DOI · Paper page
  61. 10.1109/TMI.2013.2246577 · DOI · Paper page
  62. pustina, 2017, Anatomy of an Registration Call
  63. 10.1016/j.neuroimage.2010.07.020 · DOI · Paper page
  64. 10.1016/j.media.2007.12.003 · DOI · Paper page
  65. 10.1007/978-3-030-78191-0_1 · DOI
  66. 10.1109/CVPR.2019.00874 · DOI · Paper page
  67. 10.1109/TMI.2005.852050 · DOI · Paper page
  68. chee, 2018, AIRNet: Self-supervised affine registration for 3D medical images using neural networks, arXiv 1810 02583
  69. 10.1016/j.neuroimage.2013.04.114 · DOI · Paper page
  70. 10.1016/j.media.2007.06.004 · DOI · Paper page
  71. 10.1016/j.cmpb.2009.09.002 · DOI · Paper page
  72. 10.1109/42.929618 · DOI · Paper page
  73. 10.1016/j.neuroimage.2008.10.040 · DOI · Paper page
  74. 10.1002/jmri.24542 · DOI · Paper page
  75. 10.1109/TMI.2019.2897538 · DOI · Paper page
  76. 10.1016/j.neuroimage.2007.12.025 · DOI · Paper page
  77. 10.1007/978-3-319-67558-9_24 · DOI · Paper page
  78. 10.1007/11566489_32 · DOI
  79. 10.1109/TITB.2008.917893 · DOI
  80. 10.1016/j.neuroimage.2018.09.060 · DOI · Paper page
  81. 10.1098/rstb.2001.0916 · DOI
  82. holmes, 2015, Brain genomics superstruct project initial data release with structural, functional, and behavioral measures, Data Science Journal, 2
  83. 10.1097/00004728-198701000-00003 · DOI
  84. 10.1016/j.neuroimage.2009.10.002 · DOI · Paper page

Cited by

See all 201

Metadata

Authors
Hoffmann, M., Billot, B., Greve, D. N., Iglesias, J. E., Fischl, B., & Dalca, A. V.
Venue
IEEE Transactions on Medical Imaging
Published
2022-03
Volume
41
Issue
3
Pages
543-558
ISSN
  • 0278-0062 (ISSN-L)
  • 1558-254X
OpenAlex
W3204867994
Cited by
201
References
14
FWCI
7.23
Type
journal article
Language
en
Open access
hybrid
Funders
  • Alzheimer’s Research UK (ARUK-IRG2019A-003)
  • European Research Council (ERC) under Starting Grant (677697)
  • National Institutes of Health (1R01 AG070988-01)
  • BRAIN Initiative Grant (1RF1MH123195-01, K99 HD101553, U01 AG052564, R56 AG064027, R01 AG064027, R01 AG016495, U01 MH117023, P41 EB015896, R01 EB023281, R01 EB019956, R01 NS0525851, R21 NS072652, R01 NS083534, U01 NS086625, U24 NS10059103, R01 NS105820, S10 RR023401, S10 RR019307, S10 RR023043)

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: SynthMorph: Learning Contrast-Invariant Registration Without Acquired Images. https://app.cochat.ai/papers/10.1109/tmi.2021.3116879
The paper
Hoffmann, M., Billot, B., Greve, D. N., Iglesias, J. E., Fischl, B., & Dalca, A. V. (2022). SynthMorph: Learning Contrast-Invariant Registration Without Acquired Images. IEEE Transactions on Medical Imaging. https://doi.org/10.1109/tmi.2021.3116879

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.