VEPCON: Source imaging of high-density visual evoked potentials with multi-scale brain parcellations and connectomes

Overview

The multimodal dataset VEPCON follows the BIDS standard and provides raw data of high-density EEG, structural MRI and diffusion weighted images (DWI) recorded in 20 participants.

Visual evoked potentials were recorded while participants discriminated briefly presented faces from scrambled faces (task-faces), or coherently moving stimuli from incoherent ones (task-motion). Note that raw EEG data for sub-05 (for both task-faces and task-motion) and for sub-15 (for task-motion) were discarded because of excessive motion. MRI and DWI were recorded in a separate session from the same participants.

VEPCON also contains data derivatives that follow as close as possible the BIDS derivatives specifications. It includes in particular: pre-processed EEG of single trials in each condition, behavioral measures, structural MRIs, Freesurfer 7.1.1 outputs of defaced MRIs, individual brain parcellations at 5 spatial resolutions (83 to 1015 regions), and corresponding structural connectomes based on fiber count, fiber density, average fractional anisotropy and mean diffusivity maps. In addition, Freesurfer's outputs include a bem/ folder that contains all files generated by MNE to describe the Boundary Element Model (BEM) based on Freesurfer's surfaces estimated from the original undefaced structural MRIs. Finally, VEPCON also provides EEG inverse solutions for source imaging based on individual anatomy, and Python and Matlab code for deriving time-series of activity in each brain region, at each parcellation level.

We believe this dataset can contribute to multimodal methods development, studying structure-function relations, as well as unimodal optimization of source imaging and graph analysis, among many other possibilities.

All code supporting the dataset can be found in the code/ folder.