Abstract
This paper demonstrates spherical convolutional neural networks (S-CNN) offer distinct advantages over conventional fully-connected networks (FCN) at estimating scalar parameters of tissue microstructure from diffusion MRI (dMRI). Such microstructure parameters are valuable for identifying pathology and quantifying its extent. However, current clinical practice commonly acquires dMRI data consisting of only 6 diffusion weighted images (DWIs), limiting the accuracy and precision of estimated microstructure indices. Machine learning (ML) has been proposed to address this challenge. However, existing ML-based methods are not robust to differing gradient schemes, nor are they rotation equivariant. Lack of robustness to differing gradient schemes requires a new network to be trained for each scheme, complicating the analysis of data from multiple sources. A possible consequence of the lack of rotational equivariance is that the training dataset must contain a diverse range of microstucture orientations. Here, we show spherical CNNs represent a compelling alternative that is robust to new gradient schemes as well as offering rotational equivariance. We show the latter can be leveraged to decrease the number of training datapoints required.
This work is supported by the EPSRC-funded UCL Centre for Doctoral Training in Medical Imaging (EP/L016478/1), the Department of Health’s NIHR-funded Biomedical Research Centre at UCLH and the Wellcome Trust.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Aliotta, E., Nourzadeh, H., Sanders, J., Muller, D., Ennis, D.B.: Highly accelerated, model-free diffusion tensor MRI reconstruction using neural networks. Med. Phys. 46(4), 1581–1591 (2019). https://doi.org/10.1002/mp.13400
Basser, P.J., Mattiello, J., LeBihan, D.: Estimation of the effective self-diffusion tensor from the NMR spin echo. J. Magn. Resonan. Ser. B 103(3), 247–254 (1994). https://doi.org/10.1006/jmrb.1994.1037
Bodini, B., Ciccarelli, O.: Diffusion MRI in Neurological Disorders. Diffusion MRI: From Quantitative Measurement to In vivo Neuroanatomy, 2nd edn, pp. 241–255 (2014). https://doi.org/10.1016/B978-0-12-396460-1.00011-1
Chen, G., et al.: Estimating tissue microstructure with undersampled diffusion data via graph convolutional neural networks. In: Martel, A.L., et al. (eds.) MICCAI 2020. LNCS, vol. 12267, pp. 280–290. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-59728-3_28
Cobb, O.J., et al.: Efficient generalized spherical CNNs. In: ICLR 2021 (2021). https://arxiv.org/abs/2010.11661
Cohen, T.S., Geiger, M., Köhler, J., Welling, M.: Spherical CNNs. In: ICLR 2018, January 2018. https://arxiv.org/abs/1801.10130
Elaldi, A., Dey, N., Kim, H., Gerig, G.: Equivariant spherical deconvolution: learning sparse orientation distribution functions from spherical data. In: Feragen, A., Sommer, S., Schnabel, J., Nielsen, M. (eds.) IPMI 2021. LNCS, vol. 12729, pp. 267–278. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-78191-0_21
Golkov, V., et al.: q-Space deep learning: twelve-fold shorter and model-free diffusion MRI scans. IEEE Trans. Med. Imaging 35(5), 1344–1351 (2016). https://doi.org/10.1109/TMI.2016.2551324
Jones, D.K., Horsfield, M.A., Simmons, A.: Optimal strategies for measuring diffusion in anisotropic systems by magnetic resonance imaging. Magn. Reson. Med. 42(3), 515–525 (1999). https://doi.org/10.1002/(SICI)1522-2594(199909)42:3<515::AID-MRM14>3.0.CO;2-Q
Jones, D.K.: The effect of gradient sampling schemes on measures derived from diffusion tensor MRI: a Monte Carlo study. Magn. Reson. Med. 51(4), 807–815 (2004). https://doi.org/10.1002/mrm.20033
Lin, Z., et al.: Fast learning of fiber orientation distribution function for MR tractography using convolutional neural network. Med. Phys. 46(7), 3101–3116, mp.13555 (2019). https://doi.org/10.1002/mp.13555
Park, J., et al.: DIFFnet: diffusion parameter mapping network generalized for input diffusion gradient schemes and b-values. IEEE Trans. Med. Imaging 41(2), 491–499 (2022). https://doi.org/10.1109/TMI.2021.3116298
Sedlar, S., Alimi, A., Papadopoulo, T., Deriche, R., Deslauriers-Gauthier, S.: A spherical convolutional neural network for white matter structure imaging via dMRI. In: de Bruijne, M., et al. (eds.) MICCAI 2021. LNCS, vol. 12903, pp. 529–539. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-87199-4_50
Skare, S., Hedehus, M., Moseley, M.E., Li, T.Q.: Condition number as a measure of noise performance of diffusion tensor data acquisition schemes with MRI. J. Magn. Resonan. 147(2), 340–352 (2000). https://doi.org/10.1006/jmre.2000.2209
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Goodwin-Allcock, T., McEwen, J., Gray, R., Nachev, P., Zhang, H. (2022). How Can Spherical CNNs Benefit ML-Based Diffusion MRI Parameter Estimation?. In: Cetin-Karayumak, S., et al. Computational Diffusion MRI. CDMRI 2022. Lecture Notes in Computer Science, vol 13722. Springer, Cham. https://doi.org/10.1007/978-3-031-21206-2_9
Download citation
DOI: https://doi.org/10.1007/978-3-031-21206-2_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21205-5
Online ISBN: 978-3-031-21206-2
eBook Packages: Computer ScienceComputer Science (R0)