Denis A. Engemann
  • Denis A. Engemann
  • About Me
  • Publications

On this page

  • Statistically consistent regression modeling from brain activity recoreded with M/EEG
  • Summary
  • Citation

Manifold-regression to predict from MEG/EEG brain signals without source modeling

paper
published
NeurIPS
neuroscience
M/EEG
machine learning
brain age
Riemannian geometry
spatial filtering
Author

Sabbagh, David, Pierre Ablin, Gaël Varoquaux, Alexandre Gramfort, and Denis A. Engemann.

Published

December 10, 2019

Statistically consistent regression modeling from brain activity recoreded with M/EEG

Summary

(reprinted from the article: “bigger picture”)

Magnetoencephalography and electroencephalography (M/EEG) can reveal neuronal dynamics non-invasively in real-time and are therefore appreciated methods in medicine and neuroscience. Recent advances in modeling brain-behavior relationships have highlighted the effectiveness of Riemannian geometry for summarizing the spatially correlated time-series from M/EEG in terms of their covariance. However, after artefact-suppression, M/EEG data is often rank deficient which limits the application of Riemannian concepts. In this article, we focus on the task of regression with rank-reduced covariance matrices. We study two Riemannian approaches that vectorize the M/EEG covariance between sensors through projection into a tangent space. The Wasserstein distance readily applies to rank-reduced data but lacks affine-invariance. This can be overcome by finding a common subspace in which the covariance matrices are full rank, enabling the affine-invariant geometric distance. We investigated the implications of these two approaches in synthetic generative models, which allowed us to control estimation bias of a linear model for prediction. We show that Wasserstein and geometric distances allow perfect out-of-sample prediction on the generative models. We then evaluated the methods on real data with regard to their effectiveness in predicting age from M/EEG covariance matrices. The findings suggest that the data-driven Riemannian methods outperform different sensor-space estimators and that they get close to the performance of biophysics-driven source-localization model that requires MRI acquisitions and tedious data processing. Our study suggests that the proposed Riemannian methods can serve as fundamental building-blocks for automated large-scale analysis of M/EEG.

Citation

@article{sabbagh2019manifold,
  title={Manifold-regression to predict from MEG/EEG brain signals without source modeling},
  author={Sabbagh, David and Ablin, Pierre and Varoquaux, Ga{\"e}l and Gramfort, Alexandre and Engemann, Denis A},
  journal={Advances in Neural Information Processing Systems},
  volume={32},
  year={2019}
}