Recent Publications

@article{richard-etal:20,
 archiveprefix = {arXiv},
 author = {Richard, H., Gresele, L., Hyvärinen, A., Thirion, B., Gramfort, A. and Ablin, P.},
 eprint = {2006.06635},
 journal = {ArXiv},
 title = {Modeling Shared Responses in Neuroimaging Studies through MultiView {ICA}},
 url = {https://arxiv.org/abs/2006.06635},
 volume = {abs/2006.06635},
 year = {2020}
}

@article{banville-etal:20,
 archiveprefix = {arXiv},
 author = {Banville, H., Chehab, O., Hyvärinen, A., Engemann, D. and Gramfort, A.},
 eprint = {2007.16104},
 journal = {ArXiv},
 title = {Uncovering the structure of clinical {EEG} signals with self-supervised
learning},
 url = {https://arxiv.org/abs/2007.16104},
 volume = {abs/2007.16104},
 year = {2020}
}

@inproceedings{janati-etal:20a,
 abstract = {Comparing data defined over space and time is notoriously hard. It involves quantifying both spatial and temporal variability while taking into account the chronological structure of the data. Dynamic Time Warping (DTW) computes a minimal cost alignment between time series that preserves the chronological order but is inherently blind to spatio-temporal shifts. In this paper, we propose Spatio-Temporal Alignments (STA), a new differentiable formulation of DTW that captures spatial and temporal variability. Spatial differences between time samples are captured using regularized Optimal transport. While temporal alignment cost exploits a smooth variant of DTW called soft-DTW. We show how smoothing DTW leads to alignment costs that increase quadratically with time shifts. The costs are expressed using an unbalanced Wasserstein distance to cope with observations that are not probabilities. Experiments on handwritten letters and brain imaging data confirm our theoretical findings and illustrate the effectiveness of STA as a dissimilarity for spatio-temporal data.},
 address = {Online},
 author = {Janati, H., Cuturi, M. and Gramfort, A.},
 booktitle = {AISTATS},
 editor = {Silvia Chiappa and Roberto Calandra},
 month = {26--28 Aug},
 pages = {1695--1704},
 pdf = {http://proceedings.mlr.press/v108/janati20a/janati20a.pdf},
 publisher = {PMLR},
 series = {Proceedings of Machine Learning Research},
 title = {Spatio-temporal alignments: Optimal transport through space and time},
 url = {http://proceedings.mlr.press/v108/janati20a.html},
 volume = {108},
 year = {2020}
}

@inproceedings{janati-etal:20b,
 author = {Janati, H., Cuturi, M. and Gramfort, A.},
 booktitle = {Proc. ICML 2020},
 month = {July},
 pdf = {https://proceedings.icml.cc/static/paper_files/icml/2020/1584-Paper.pdf},
 title = {Debiased Sinkhorn Barycenters},
 year = {2020}
}

@inproceedings{bertrand-etal:20,
 author = {Bertrand, Q., Klopfenstein, Q., Blondel, M., Vaiter, S., Gramfort, A. and Salmon, J.},
 booktitle = {Proc. ICML 2020},
 month = {July},
 pdf = {https://proceedings.icml.cc/static/paper_files/icml/2020/1831-Paper.pdf},
 title = {Implicit differentiation of Lasso-type models for hyperparameter optimization},
 url = {https://arxiv.org/abs/2002.08943},
 year = {2020}
}

@article{engemann-etal:20,
 abstract = {Electrophysiological methods, that is M/EEG, provide unique views into brain health. Yet, when building predictive models from brain data, it is often unclear how electrophysiology should be combined with other neuroimaging methods. Information can be redundant, useful common representations of multimodal data may not be obvious and multimodal data collection can be medically contraindicated, which reduces applicability. Here, we propose a multimodal model to robustly combine MEG, MRI and fMRI for prediction. We focus on age prediction as a surrogate biomarker in 674 subjects from the Cam-CAN dataset. Strikingly, MEG, fMRI and MRI showed additive effects supporting distinct brain-behavior associations. Moreover, the contribution of MEG was best explained by cortical power spectra between 8 and 30 Hz. Finally, we demonstrate that the model preserves benefits of stacking when some data is missing. The proposed framework, hence, enables multimodal learning for a wide range of biomarkers from diverse types of brain signals.},
 author = {Engemann, D., Kozynets, O., Sabbagh, D., Lemaître, G., Varoquaux, G., Liem, F. and Gramfort, A.},
 citation = {eLife 2020;9:e54055},
 doi = {10.7554/eLife.54055},
 editor = {Shackman, Alexander and de Lange, Floris P and Tsvetanov, Kamen and Trujillo-Barreto, Nelson},
 issn = {2050-084X},
 journal = {eLife},
 keywords = {biomarker, aging, magnetic resonance imaging, magnetoencephalogrphy, oscillations, machine learning},
 month = {may},
 pages = {e54055},
 pub_date = {2020-05-19},
 publisher = {eLife Sciences Publications, Ltd},
 title = {Combining magnetoencephalography with magnetic resonance imaging enhances learning of surrogate-biomarkers},
 url = {https://doi.org/10.7554/eLife.54055},
 volume = {9},
 year = {2020}
}

@article{sabbagh-etal:20,
 abstract = {Predicting biomedical outcomes from Magnetoencephalography and Electroencephalography (M/EEG) is central to applications like decoding, brain-computer-interfaces (BCI) or biomarker development and is facilitated by supervised machine learning. Yet, most of the literature is concerned with classification of outcomes defined at the event-level. Here, we focus on predicting continuous outcomes from M/EEG signal defined at the subject-level, and analyze about 600 MEG recordings from Cam-CAN dataset and about 1000 EEG recordings from TUH dataset. Considering different generative mechanisms for M/EEG signals and the biomedical outcome, we propose statistically-consistent predictive models that avoid source-reconstruction based on the covariance as representation. Our mathematical analysis and ground-truth simulations demonstrated that consistent function approximation can be obtained with supervised spatial filtering or by embedding with Riemannian geometry. Additional simulations revealed that Riemannian methods were more robust to model violations, in particular geometric distortions induced by individual anatomy. To estimate the relative contribution of brain dynamics and anatomy to prediction performance, we propose a novel model inspection procedure based on biophysical forward modeling. Applied to prediction of outcomes at the subject-level, the analysis revealed that the Riemannian model better exploited anatomical information while sensitivity to brain dynamics was similar across methods. We then probed the robustness of the models across different data cleaning options. Environmental denoising was globally important but Riemannian models were strikingly robust and continued performing well even without preprocessing. Our results suggest each method has its niche: supervised spatial filtering is practical for event-level prediction while the Riemannian model may enable simple end-to-end learning.},
 author = {Sabbagh, D., Ablin, P., Varoquaux, G., Gramfort, A. and Engemann, D.},
 doi = {https://doi.org/10.1016/j.neuroimage.2020.116893},
 issn = {1053-8119},
 journal = {NeuroImage},
 keywords = {MEG/EEG, Neuronal oscillations, Machine learning, Covariance, Spatial filters, Riemannian geometry},
 pages = {116893},
 title = {Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states},
 url = {http://www.sciencedirect.com/science/article/pii/S1053811920303797},
 volume = {222},
 year = {2020}
}

Full list of publications

Short Bio

I am currently senior research scientist (DR, HDR) at Inria in the Parietal Team. My work is on statistical machine learning, signal and image processing, optimization, scientific computing and software engineering with primary applications in brain functional imaging (MEG, EEG, fMRI). Before joining Inria, I was an assistant professor for 5 years at Telecom ParisTech in the signal processing and machine learning department and before I was at the Martinos Center for Biomedical Imaging at Harvard University in Boston. I am also a co-director of the Center for Data Science at Université Paris-Saclay. In 2015, I got an ERC Starting Grant for my project called Signal and Learning Applied to Brain data (SLAB) and in 2019 a grant ANR Chaire on Arficial Intelligence called BrAIN.

Contact

Email: alexandre.gramfort@inria.fr

Address: Inria Saclay Île-de-France, Bâtiment Alan Turing, 1 rue Honoré d'Estienne d'Orves, Campus de l'École Polytechnique 91120 Palaiseau

Software

  • scikit-learn - A Python project for machine learning.
  • MNE - A complete package to process EEG and MEG data: forward and inverse problems, preprocessing, stats, time-frequency analysis.
  • openmeeg - C++ package for low-frequency bio-electromagnetism including the EEG/MEG forward problem. OpenMEEG implements the Symmetric BEM which has shown to provide very accurate solutions. Some features: parallel processing, Python Bindings, Matlab integration with Fieldtrip and BrainStorm.

More on my Github Page

Team

Engineers

Post-docs

PhD Students

Alumni

Positions

  • PhD/Post-doc positions on machine learning and signal processing with applications in neuroimaging (MEG, EEG)

This list is fuzzy so please contact me directly for potential opportunities.

Teaching