Abstract
Age predictions from T1-weighted (T1w) MR brain images using deep learning models have become increasingly more accurate, mainly with the construction of larger and more complex model architectures, such as cascade networks, but also with the use of larger training datasets. We adopted and evaluated a data augmentation strategy called Mixup that combines input T1w brain scans and associated output ages for the brain age regression task. On a multi-site dataset of 2504 T1w brain scans we evaluated and tested multiple mixing factor distributions, applied mixing of similar/different sample pairs based on low/high age difference, and combined mixing in auxiliary variables. We found consistent improvements in prediction accuracy with the use of Mixup augmentation, with minimal computational overhead, and, despite using a simple VGG-based deep learning model architecture, achieved a highly competitive mean absolute error as low as 2.96 years.
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
Notes
- 1.
Adaptive non-local means denoising: https://github.com/djkwon/naonlm3d.
- 2.
NiftyReg registration: http://cmictig.cs.ucl.ac.uk/wiki/index.php/NiftyReg.
- 3.
N4 bias field correction: https://manpages.debian.org/testing/ants/N4BiasFieldCorrection.1.en.html.
References
Bron, E.E., et al.: Cross-cohort generalizability of deep and conventional machine learning for MRI-based diagnosis and prediction of Alzheimer’s disease. NeuroImage: Clin. 31, 102712 (2021). https://doi.org/10.1016/j.nicl.2021.102712. https://www.sciencedirect.com/science/article/pii/S221315822100156X
Cheng, J., et al.: Brain age estimation from mri using cascade networks with ranking loss. In: IEEE Transactions on Medical Imaging, pp. 1–1 (2021). https://doi.org/10.1109/TMI.2021.3085948
Cole, J.H., et al.: Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker. Neuroimage 163, 115–124 (2017). https://doi.org/10.1016/j.neuroimage.2017.07.059. http://www.sciencedirect.com/science/article/pii/S1053811917306407
Eaton-Rosen, Z., Bragman, F., Ourselin, S., Cardoso, M.J.: Improving data augmentation for medical image segmentation. In: International Conference on Medical Imaging with Deep Learning, p. 3 (2018)
Fonov, V., Evans, A., McKinstry, R., Almli, C., Collins, D.: Unbiased nonlinear average age-appropriate brain templates from birth to adulthood. Neuroimage 47, S102 (2009). https://doi.org/10.1016/S1053-8119(09)70884-5. https://www.sciencedirect.com/science/article/pii/S1053811909708845
Galdran, A., Carneiro, G., Ballester, M.A.G.: Balanced-MixUp for highly imbalanced medical image classification. arXiv:2109.09850 [cs] (2021). http://arxiv.org/abs/2109.09850, arXiv: 2109.09850
Hwang, S.H., Whang, S.E.: MixR: data mixing augmentation for regression (2021). https://doi.org/10.48550/ARXIV.2106.03374. https://arxiv.org/abs/2106.03374
Isaksson, L.J., et al.: Mixup (sample pairing) can improve the performance of deep segmentation networks. J. Artif. Intell. Soft Comput. Res. 12(1), 29–39 (2022). https://doi.org/10.2478/jaiscr-2022-0003. https://www.sciendo.com/article/10.2478/jaiscr-2022-0003
Manjón, J.V., Coupé, P., Martí-Bonmatí, L., Collins, D.L., Robles, M.: Adaptive non-local means denoising of MR images with spatially varying noise levels. J. Magn. Reson. Imaging 31(1), 192–203 (2010). https://doi.org/10.1002/jmri.22003
Marcus, D.S., Fotenos, A.F., Csernansky, J.G., Morris, J.C., Buckner, R.L.: Open access series of imaging studies: longitudinal MRI data in nondemented and demented older adults. J. Cogn. Neurosci. 22(12), 2677–2684 (2010). https://doi.org/10.1162/jocn.2009.21407
Modat, M., Cash, D.M., Daga, P., Winston, G.P., Duncan, J.S., Ourselin, S.: Global image registration using a symmetric block-matching approach. J. Med. Imaging 1(2), 1–6 (2014). https://doi.org/10.1117/1.JMI.1.2.024003. https://doi.org/10.1117/1.JMI.1.2.024003
Panfilov, E., Tiulpin, A., Klein, S., Nieminen, M.T., Saarakkala, S.: Improving robustness of deep learning based knee MRI segmentation: Mixup and adversarial domain adaptation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops (2019)
Shafto, M.A., et al.: The Cambridge centre for ageing and neuroscience (Cam-CAN) study protocol: a cross-sectional, lifespan, multidisciplinary examination of healthy cognitive ageing. BMC Neurol. 14 (2014). https://doi.org/10.1186/s12883-014-0204-1. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4219118/
Taylor, J.R., et al.: The Cambridge centre for ageing and neuroscience (Cam-CAN) data repository: structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample. Neuroimage 144(Pt B), 262–269 (2017). https://doi.org/10.1016/j.neuroimage.2015.09.018
The Cambridge Centre for Ageing and Neuroscience (CamCAN). http://fcon_1000.projects.nitrc.org/indi/abide/abide_I.html
Tustison, N.J., et al.: N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging 29(6), 1310–1320 (2010). https://doi.org/10.1109/TMI.2010.2046908
Verma, V., et al.: Manifold Mixup: better representations by interpolating hidden states. In: Proceedings of the 36th International Conference on Machine Learning, pp. 6438–6447. PMLR (2019). https://proceedings.mlr.press/v97/verma19a.html
Xie, T., Cheng, X., Liu, M., Deng, J., Wang, X., Liu, M.: Thumbnail: a novel data augmentation for convolutional neural network. arXiv:2103.05342 [cs] (2021). http://arxiv.org/abs/2103.05342. arXiv: 2103.05342
Yun, S., Han, D., Oh, S.J., Chun, S., Choe, J., Yoo, Y.: CutMix: regularization strategy to train strong classifiers with localizable features. arXiv:1905.04899 [cs] (2019). http://arxiv.org/abs/1905.04899. arXiv: 1905.04899
Zhang, H., Cisse, M., Dauphin, Y.N., Lopez-Paz, D.: mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 (2017)
Acknowledgements
This study was supported by the Slovenian Research Agency (Core Research Grant No. P2-0232 and Research Grants Nos. J2-2500 and J2-3059).
Data collection and sharing for this project was partially provided by:
– Alzheimer’s Disease Neuroimaging Initiative (ADNI) database
(adni.loni.usc.edu). The investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.
– Cambridge Centre for Ageing and Neuroscience (CamCAN). CamCAN funding was provided by the UK Biotechnology and Biological Sciences Research Council (grant number BB/H008217/1), together with support from the UK Medical Research Council and University of Cambridge, UK.
– OASIS Longitudinal. Principal Investigators: D. Marcus, R, Buckner, J. Csernansky, J. Morris; P50 AG05681, P01 AG03991, P01 AG026276, R01 AG021910, P20 MH071616, U24 RR021382.
– ABIDE I. Primary support for the work by Adriana Di Martino was provided by the (NIMH K23MH087770) and the Leon Levy Foundation.
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
Dular, L., Špiclin, Ž. (2022). Mixup Augmentation Improves Age Prediction from T1-Weighted Brain MRI Scans. In: Rekik, I., Adeli, E., Park, S.H., Cintas, C. (eds) Predictive Intelligence in Medicine. PRIME 2022. Lecture Notes in Computer Science, vol 13564. Springer, Cham. https://doi.org/10.1007/978-3-031-16919-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-16919-9_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16918-2
Online ISBN: 978-3-031-16919-9
eBook Packages: Computer ScienceComputer Science (R0)
