English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Preprint

Towards the interpretability of deep learning models for human neuroimaging

MPS-Authors
/persons/resource/persons227300

Hofmann,  Simon       
Department Neurology, MPI for Human Cognitive and Brain Sciences, Max Planck Society;
Department of Artificial Intelligence, Fraunhofer Institute Heinrich Hertz;
Clinic for Cognitive Neurology, University of Leipzig Medical Center;

/persons/resource/persons225526

Beyer,  Frauke
Department Neurology, MPI for Human Cognitive and Brain Sciences, Max Planck Society;
Department of Artificial Intelligence, Fraunhofer Institute Heinrich Hertz;

/persons/resource/persons20065

Villringer,  Arno
Department Neurology, MPI for Human Cognitive and Brain Sciences, Max Planck Society;
Clinic for Cognitive Neurology, University of Leipzig Medical Center;
MindBrainBody Institute, Berlin School of Mind and Brain, Humboldt-Universität zu Berlin;
Center for Stroke Research, Charité – Universitätsmedizin Berlin;

/persons/resource/persons128137

Witte,  A. Veronica
Department Neurology, MPI for Human Cognitive and Brain Sciences, Max Planck Society;
Clinic for Cognitive Neurology, University of Leipzig Medical Center;

External Resource
No external resources are shared
Fulltext (restricted access)
There are currently no full texts shared for your IP range.
Fulltext (public)

Hofmann_pre.pdf
(Preprint), 9MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Hofmann, S., Beyer, F., Lapuschkin, S., Loeffler, M., Müller, K.-R., Villringer, A., et al. (2021). Towards the interpretability of deep learning models for human neuroimaging. bioRxiv. doi:10.1101/2021.06.25.449906.


Cite as: https://hdl.handle.net/21.11116/0000-0009-9FF8-D
Abstract
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 aging at both small and large-scale changes, revealing gross enlargements of ventricles and subarachnoid spaces, as well as lesions, iron accumulations and atrophies that appear throughout the brain. Divergence from expected aging reflected cardiovascular risk
factors and accelerated aging was more pronounced in the frontal lobe. Applying LRP, our study demonstrates how superior deep learning models detect brain-aging in healthy and at-risk individuals throughout adulthood.