Yohan Chatelain

The paper quantifies numerical variability in MRI analyses of Parkinson's disease and shows it can significantly affect statistical conclusions and p-values.

The paper shows that untrained convolutional neural networks can extract features from structural MRI data that perform as well as or better than pretrained models in three tasks across three datasets.

The paper evaluates numerical variability in functional MRI graph measures using the fMRIPrep pipeline and finds measurable impact on network outcomes.

This study examines how different versions of FreeSurfer affect structural brain measure estimation in Parkinson's disease and shows significant software-induced variability impacting clinical outcomes.

This study replicates findings from Hanganu et al. (2014) on Parkinson's disease neuroimaging biomarkers using PPMI data, finding partial replication of cortical thinning and cognitive decline associations.

This study attempts to reproduce and replicate a machine learning model for predicting Parkinson's disease progression using MRI data, finding lower performance than the original study.

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.

The paper uses Monte Carlo Arithmetic to perturb structural connectome pipelines, creating augmented datasets that improve age classification performance across multiple classifiers and preprocessing strategies.