Alexandre Gramfort

EMG-BIDS standardizes EMG data organization, enabling FAIR data sharing and supporting machine learning applications.

This study evaluates how robust different BSI methods are to noise level misspecification using simulations and metrics like EMD.

The paper investigates how working memory stores durations, finding that the number of durations, not their total length, affects recall precision and is reflected in alpha and beta oscillations.

This paper provides guidelines for applying simulation based inference to estimate parameters in large-scale biophysically detailed neural models, focusing on time series data like MEG/EEG waveforms.

This study predicts age from at-home EEG data during sleep and meditation, achieving R2 scores of 0.3-0.5, comparable to lab-grade EEG benchmarks.

This paper improves M/EEG-based regression models by applying covariance-based domain adaptation techniques to address dataset shifts across different recording contexts.

This paper benchmarks machine learning approaches for brain age prediction using M/EEG signals across four international cohorts, finding that spatially aware models achieve the highest performance.

This study compares the predictive power of oxygenated and de-oxygenated hemoglobin signals in fNIRS data for experimental condition classification, finding oxygenated hemoglobin provides better differentiation.

Across 101 listeners, how well GPT-2 layer-8 activations predict fMRI responses to spoken stories tracks story comprehension, with a correlation of R = 0.50 across subject-story pairs.

The paper predicts future cognitive decline using non-brain and multimodal brain imaging data in healthy and pathological aging, finding that non-brain data and structural MRI improve predictions.

This paper develops predictive models for continuous biomedical outcomes from M/EEG data, comparing supervised spatial filtering and Riemannian geometry approaches, and finds Riemannian methods are more robust to anatomical distortions.

The paper combines MEG, MRI, and fMRI to predict age as a surrogate biomarker, showing additive effects and demonstrating robustness to missing data.

This study compares LCMV beamformer implementations across four MEG toolboxes, finding reliable source localization at typical SNR but differences in sensitivity to preprocessing parameters and SNR effects.

The paper investigates how phase-amplitude coupling in the alpha-beta band affects the precision of self-generated timing tasks.

The paper reanalyzes an open MEG/EEG dataset using MNE to compare different analysis strategies and provide best practices for group studies.

MEG-BIDS extends the BIDS standard to organize MEG data, enabling efficient storage, sharing, and integration with MRI data.

This paper introduces non-linear auto-regressive models to detect and quantify cross-frequency coupling in neural time series, showing improved performance over traditional methods.

This study decodes brain activity during visual motion coherence discrimination using MEG, finding that perceptual thresholds align with decoder boundaries and identifying the earliest categorization in hMT+.