Abstract
Clinical MR imaging is typically qualitative, i.e. the observed signal reflects the underlying tissue contrast, but the measurements are not meaningful when taken in isolation. Quantitative MR imaging maps are rarely acquired due to time and complexity constraints but directly measure intrinsic tissue properties, allowing for explicit tissue characterisation and MR contrast simulation. A machine learning network trained on quantitative MRI would circumvent the need to design contrast-agnostic models, minimise domain shift issues, and reduce complex data pre-processing. Such models would also be future-proof by design, anticipating new qualitative sequence developments and changes in acquisition parameters. In this work, we propose a new Bloch-equation-based physics-informed unsupervised network that learns to map qualitative MR data to their quantitative equivalents without the need for paired qualitative-quantitative data. Furthermore, we make the proposed model robust to missing data, enabling us to map any arbitrary set of qualitative data from a patient into quantitative Multi-Parametric Maps (MPMs). We demonstrate that the estimated MPMs are a robust and invariant data representation, are self-consistent, enable missing data imputation, and facilitate data harmonisation from multiple sites while bridging algorithmic domain gaps.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Project MONAI. DOI: 10.5281/zenodo.4323059
Abd-Ellah, M.K., Awad, A.I., Khalaf, A.A.M., Hamed, H.F.A.: A review on brain tumor diagnosis from MRI images: practical implications, key achievements, and lessons learned. Magn. Reson. Imaging 61, 300–318 (2019). https://doi.org/10.1016/j.mri.2019.05.028
Andreasen, N.C., et al.: T1 and T2 relaxation times in schizophrenia as measured with magnetic resonance imaging. Schizophr. Res. 5(3), 223–232 (1991). https://doi.org/10.1016/0920-9964(91)90080-B
Bandettini, P.A.: Twenty years of functional MRI: the science and the stories. Neuroimage 62(2), 575–588 (2012). https://doi.org/10.1016/j.neuroimage.2012.04.026
Baudrexel, S., et al.: Quantitative mapping of T1 and T2* discloses nigral and brainstem pathology in early Parkinson’s disease. Neuroimage 51(2), 512–520 (2010). https://doi.org/10.1016/j.neuroimage.2010.03.005
Borges, P., et al.: Physics-informed brain MRI segmentation. 11827 LNCS, 100–109 (2020). https://doi.org/10.1007/978-3-030-32778-1_11
Brudfors, M., Balbastre, Y., Nachev, P., Ashburner, J.: MRI Super-Resolution using Multi-Channel Total Variation. https://brain-development.org/ixi-dataset/
Brudfors, M., Balbastre, Y., Nachev, P., Ashburner, J.: A Tool for Super-Resolving Multimodal Clinical MRI (2019). 10.48550/arxiv.1909.01140. https://arxiv.org/abs/1909.01140v1
Cai, C., et al.: Single-shot T2 mapping using overlapping-echo detachment planar imaging and a deep convolutional neural network. Magn. Reson. Med. 80(5), 2202–2214 (2018). https://doi.org/10.1002/MRM.27205. https://onlinelibrary.wiley.com/doi/full/10.1002/mrm.27205
Cardoso, M.J., et al.: Geodesic information flows: spatially-variant graphs and their application to segmentation and fusion. IEEE Trans. Med. Imaging 34(9), 1976–1988 (2015)
Chavhan, G.B.: Appropriate selection of MRI sequences for common scenarios in clinical practice. Pediatr. Radiol. 46(6), 740–747 (2016). https://doi.org/10.1007/s00247-016-3556-4
Cohen, O., Zhu, B., Rosen, M.S.: MR fingerprinting Deep RecOnstruction NEtwork (DRONE). Magn. Reson. Med. 80(3), 885–894 (2018). https://doi.org/10.1002/MRM.27198, https://onlinelibrary.wiley.com/doi/full/10.1002/mrm.27198
Deichmann, R., Good, C.D., Josephs, O., Ashburner, J., Turner, R.: Optimization of 3-D MP-RAGE sequences for structural brain imaging. NeuroImage 12(1), 112–127 (2000). https://doi.org/10.1006/NIMG.2000.0601,https://pubmed.ncbi.nlm.nih.gov/10875908/
Desikan, R., et al.: Automated MRI measures identify individuals with mild cognitive impairment and Alzheimer’s disease. Brain : J. Neurol. 132(Pt 8), 2048–2057 (2009). https://doi.org/10.1093/BRAIN/AWP123, https://pubmed.ncbi.nlm.nih.gov/19460794/
Fang, Z., et al.: Deep learning for fast and spatially constrained tissue quantification from highly accelerated data in magnetic resonance fingerprinting. IEEE Trans. Med. Imaging 38(10), 2364–2374 (2019). https://doi.org/10.1109/TMI.2019.2899328
Hornak, J.: The Basics of MRI. https://www.cis.rit.edu/htbooks/mri/
Jia, X., Liu, Y., Yang, Z., Yang, D.: Multi-modality self-attention aware deep network for 3D biomedical segmentation. BMC Medical Informatics and Decision Making 20(3), 1–7 (jul 2020). https://doi.org/10.1186/s12911-020-1109-0
Jog, A., et al.: MR image synthesis by contrast learning on neighborhood ensembles. Med. Image Anal. 24(1), 63–76 (2015)
Marcus, D.S., Wang, T.H., Parker, J., Csernansky, J.G., Morris, J.C., Buckner, R.L.: Open access series of imaging studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults. J. Cogn. Neurosci. 19(9), 1498–1507 (2007). https://doi.org/10.1162/JOCN.2007.19.9.1498
Protti, A., et al.: Development and validation of a new MRI simulation technique that can reliably estimate optimal in vivo scanning parameters in a glioblastoma murine model. PloS ONE 13(7) (2018). https://doi.org/10.1371/JOURNAL.PONE.0200611, https://pubmed.ncbi.nlm.nih.gov/30036367/
Rohrer, J.D., et al.: Presymptomatic cognitive and neuroanatomical changes in genetic frontotemporal dementia in the Genetic Frontotemporal dementia Initiative (GENFI) study: a cross-sectional analysis. Articles Lancet Neurol. 14, 253–62 (2015). https://doi.org/10.1016/S1474-4422(14)70324-2, https://dx.doi.org/10.1016/
Sudlow, C., et al.: UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLOS Med. 12(3), e1001779 (2015). https://doi.org/10.1371/JOURNAL.PMED.1001779, https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1001779
Tillin, T., et al.: Southall and Brent Revisited: Cohort profile of SABRE, a UK population-based comparison of cardiovascular disease and diabetes in people of European, Indian Asian and African Caribbean origins. Int. J. epidemiol. 41(1), 33–42 (2012). https://doi.org/10.1093/IJE/DYQ175, https://pubmed.ncbi.nlm.nih.gov/21044979/
Ulyanov, D., Vedaldi, A., Lempitsky, V., Ulyanov, D., Vedaldi, A., Lempitsky, V.: Deep image prior. Int. J. Comput. Vis. 128(7), 1867–1888 (2017). https://doi.org/10.1007/s11263-020-01303-4, arXiv:1711.10925v4
Varadarajan, D., Bouman, K.L., van der Kouwe, A., Fischl, B., Dalca, A.V.: Unsupervised learning of MRI tissue properties using MRI physics models (2021). arXiv.org:2107.02704v1
Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 586–595 (2018). https://doi.org/10.1109/CVPR.2018.00068
Zhu, B., Liu, J.Z., Cauley, S.F., Rosen, B.R., Rosen, M.S.: Image reconstruction by domain-transform manifold learning. Nature 2018 555:7697 555(7697), 487–492 (2018). https://doi.org/10.1038/nature25988, https://www.nature.com/articles/nature25988
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Borges, P., Fernandez, V., Tudosiu, P.D., Nachev, P., Ourselin, S., Cardoso, M.J. (2023). Unsupervised Heteromodal Physics-Informed Representation of MRI Data: Tackling Data Harmonisation, Imputation and Domain Shift. In: Wolterink, J.M., Svoboda, D., Zhao, C., Fernandez, V. (eds) Simulation and Synthesis in Medical Imaging. SASHIMI 2023. Lecture Notes in Computer Science, vol 14288. Springer, Cham. https://doi.org/10.1007/978-3-031-44689-4_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-44689-4_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44688-7
Online ISBN: 978-3-031-44689-4
eBook Packages: Computer ScienceComputer Science (R0)