Learning patterns of the ageing brain in MRI using deep convolutional networks

NeuroImage 2021
Nicola Dinsdale1
Emma Bluemke1
Steve Smith1
Zobair Ayra1
Diego Vidaurre1,3
Mark Jenkinson1
Ana Namburete2

1Wellcome Centre for Integrative Neuroimaging, University of Oxford
2Ultrasound NeuroImage Analysis Group, University of Oxford
3Department of Psychiatry, University of Oxford

[Paper]
[GitHub]

Abstract

Both normal ageing and neurodegenerative diseases cause morphological changes to the brain. Age-related brain changes are subtle, nonlinear, and spatially and temporally heterogenous, both within a subject and across a population. Machine learning models are particularly suited to capture these patterns and can produce a model that is sensitive to changes of interest, despite the large variety in healthy brain appearance. In this paper, the power of convolutional neural networks (CNNs) and the rich UK Biobank dataset, the largest database currently available, are harnessed to address the problem of predicting brain age. We developed a 3D CNN architecture to predict chronological age, using a training dataset of 12,802 T1-weighted MRI images and a further 6,885 images for testing. The proposed method shows competitive performance on age prediction, but, most importantly, the CNN prediction errors correlated significantly with many clinical measurements from the UK Biobank in the female and male groups. In addition, having used images from only one imaging modality in this experiment, we examined the relationship between ΔBrainAge and the image-derived phenotypes (IDPs) from all other imaging modalities in the UK Biobank, showing correlations consistent with known patterns of ageing. Furthermore, we show that the use of nonlinearly registered images to train CNNs can lead to the network being driven by artefacts of the registration process and missing subtle indicators of ageing, limiting the clinical relevance. Due to the longitudinal aspect of the UK Biobank study, in the future it will be possible to explore whether the ΔBrainAge from models such as this network were predictive of any health outcomes.

Pipeline
We train an ensemble of CNNs to regress age from the T1 weighted MR images using subjects from the UK Biobank. We then deconfound for age and explore the correlations with the IDPs and non-imaging variables from the the UK Biobank.

 [GitHub]

Results

Correlations with UK Biobank Variables

We find biologically meaningful correlations with both the lifestyle factors and the image derived parameters from the UK Biobank, which follow our expectation of ageing from across the literature.


Effect of Registration

We also show that the registration used in the preprocessing pipeline affects the model prediction. Saliency and attention maps show that non-linear registration causes the model to be drive by registration artefacts...

...and that this therefore can suppress important information about the ageing brain.

Acknowledgements

This work was supported in parts by the Engineering and Physical Sciences Research Council (EPSRC) and Medical Research Council (MRC) [grant number EP/L016052/1] (N.D, E.B. and Z.A.), the Clarendon fund (E.B), and a Wellcome Trust Strategic Award (098369/Z/12/Z) (S.S. and D.V.). A.N. is grateful for support from the UK Royal Academy of Engineering under the Engineering for Development Research Fellowships scheme. M.J. is supported by the National Institute for Health Research (NIHR) and the Oxford Biomedical Research Centre (BRC). The Wellcome Centre for Integrative Neuroimaging is supported by core funding from the Wellcome Trust (203139/Z/16/Z). Computation used the Oxford Biomedical Research Computing (BMRC) facility, a joint development between the Wellcome Centre for Human Genetics and the Big Data Institute supported by Health Data Research UK and the NIHR Oxford Biomedical Research Centre. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. This research has been conducted in part using the UK Biobank Resource under Application Number 8107. We are grateful to UK Biobank for making the data available, and to all UK Biobank study participants, who generously donated their time to make this resource possible. Analysis was carried out on the clusters at the Oxford Biomedical Research Computing (BMRC) facility and FMRIB (part of the Wellcome Centre for Integrative Neuroimaging). BMRC is a joint development between the Wellcome Centre for Human Genetics and the Big Data Institute, supported by Health Data Research UK and the NIHR Oxford Biomedical Research Centre. This template was originally made by Phillip Isola and Richard Zhang for a colorful ECCV project; the code can be found here.