Towards the interpretability of deep learning models for multi-modal neuroimaging: Finding structural changes of the ageing brain
Standard
Towards the interpretability of deep learning models for multi-modal neuroimaging: Finding structural changes of the ageing brain. / Hofmann, Simon; Beyer, Frauke; Lapuschkin, Sebastian; Goltermann, Ole; Loeffler, Markus; Müller, Klaus-Robert; Villringer, Arno; Samek, Wojciech; Witte, A Veronica.
In: NEUROIMAGE, Vol. 261, 119504, 01.11.2022, p. 119504.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Towards the interpretability of deep learning models for multi-modal neuroimaging: Finding structural changes of the ageing brain
AU - Hofmann, Simon
AU - Beyer, Frauke
AU - Lapuschkin, Sebastian
AU - Goltermann, Ole
AU - Loeffler, Markus
AU - Müller, Klaus-Robert
AU - Villringer, Arno
AU - Samek, Wojciech
AU - Witte, A Veronica
PY - 2022/11/1
Y1 - 2022/11/1
N2 - Brain-age (BA) estimates based on deep learning are increasingly used as neuroimaging biomarker for brain health; however, the underlying neural features have remained unclear. We combined ensembles of convolutional neural networks with Layer-wise Relevance Propagation (LRP) to detect which brain features contribute to BA. Trained on magnetic resonance imaging (MRI) data of a population-based study (n = 2637, 18–82 years), our models estimated age accurately based on single and multiple modalities, regionally restricted and whole-brain images (mean absolute errors 3.37–3.86 years). We find that BA estimates capture ageing at both small and large-scale changes, revealing gross enlargements of ventricles and subarachnoid spaces, as well as white matter lesions, and atrophies that appear throughout the brain. Divergence from expected ageing reflected cardiovascular risk factors and accelerated ageing was more pronounced in the frontal lobe. Applying LRP, our study demonstrates how superior deep learning models detect brain-ageing in healthy and at-risk individuals throughout adulthood.
AB - Brain-age (BA) estimates based on deep learning are increasingly used as neuroimaging biomarker for brain health; however, the underlying neural features have remained unclear. We combined ensembles of convolutional neural networks with Layer-wise Relevance Propagation (LRP) to detect which brain features contribute to BA. Trained on magnetic resonance imaging (MRI) data of a population-based study (n = 2637, 18–82 years), our models estimated age accurately based on single and multiple modalities, regionally restricted and whole-brain images (mean absolute errors 3.37–3.86 years). We find that BA estimates capture ageing at both small and large-scale changes, revealing gross enlargements of ventricles and subarachnoid spaces, as well as white matter lesions, and atrophies that appear throughout the brain. Divergence from expected ageing reflected cardiovascular risk factors and accelerated ageing was more pronounced in the frontal lobe. Applying LRP, our study demonstrates how superior deep learning models detect brain-ageing in healthy and at-risk individuals throughout adulthood.
U2 - 10.1016/j.neuroimage.2022.119504
DO - 10.1016/j.neuroimage.2022.119504
M3 - SCORING: Journal article
C2 - 35882272
VL - 261
SP - 119504
JO - NEUROIMAGE
JF - NEUROIMAGE
SN - 1053-8119
M1 - 119504
ER -