An Uncertainty-Aware, Shareable and Transparent Neural Network Architecture for Brain-Age Modeling
Abstract
The deviation between chronological age and age predicted from neuroimaging data has been identified as a sensitive risk-marker of cross-disorder brain changes, growing into a cornerstone of biological age-research. However, Machine Learning models underlying the field do not consider uncertainty, thereby confounding results with training data density and variability. Also, existing models are commonly based on homogeneous training sets, often not independently validated, and cannot be shared due to data protection issues. Here, we introduce an uncertainty-aware, shareable, and transparent Monte-Carlo Dropout Composite-Quantile-Regression (MCCQR) Neural Network trained on N=10,691 datasets from the German National Cohort. The MCCQR model provides robust, distribution-free uncertainty quantification in high-dimensional neuroimaging data, achieving lower error rates compared to existing models across ten recruitment centers and in three independent validation samples (N=4,004). In two examples, we demonstrate that it prevents spurious associations and increases power to detect accelerated brain-aging. We make the pre-trained model publicly available.
Cite
@article{arxiv.2107.07977,
title = {An Uncertainty-Aware, Shareable and Transparent Neural Network Architecture for Brain-Age Modeling},
author = {Tim Hahn and Jan Ernsting and Nils R. Winter and Vincent Holstein and Ramona Leenings and Marie Beisemann and Lukas Fisch and Kelvin Sarink and Daniel Emden and Nils Opel and Ronny Redlich and Jonathan Repple and Dominik Grotegerd and Susanne Meinert and Jochen G. Hirsch and Thoralf Niendorf and Beate Endemann and Fabian Bamberg and Thomas Kröncke and Robin Bülow and Henry Völzke and Oyunbileg von Stackelberg and Ramona Felizitas Sowade and Lale Umutlu and Börge Schmidt and Svenja Caspers and German National Cohort Study Center Consortium and Harald Kugel and Tilo Kircher and Benjamin Risse and Christian Gaser and James H. Cole and Udo Dannlowski and Klaus Berger},
journal= {arXiv preprint arXiv:2107.07977},
year = {2021}
}