Gael P. Varoquaux

This paper investigates how numerical uncertainty in brain imaging pipelines affects connectome stability and shows that induced variance can improve network reliability and classification of individual differences.

This paper uses machine learning on brain imaging and sociodemographic data from the UK Biobank to create proxy measures of psychological constructs, which are as useful or more useful than traditional assessments for predicting health behaviors.

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.

The paper uses machine learning to rank cognitive domains based on brain data to predict schizophrenia in a multi-site dataset.

The paper compares statistical inference and prediction methods in biomedical data, showing they often diverge in performance and interpretation.

This paper introduces BIDS Apps, a framework for creating portable neuroimaging analysis tools compatible with BIDS data, improving reproducibility and ease of use.

This study predicts brain-age from multimodal imaging data and finds it captures cognitive impairment with robustness to motion and generalizability across datasets.