IBC

Overview
========
Functional Magnetic Resonance Imaging (fMRI) has opened the door
to brain mapping of perceptual, motor, or cognitive functions. As such, it
provides an instrumental basis for the whole field of cognitive neuroimag-
ing. However, there exists to date no data collection that systematically
maps representations for a wide-variety of mental functions at a fine spa-
tial scale. The Individual Brain Charting (IBC) project is collecting a
high-resolution multi-task-fMRI dataset, to provide an objective basis for
a comprehensive atlas of brain responses. The data refer to a cohort of
twelve(*) participants performing many different tasks. Acquiring a large
amount of tasks on the same subjects yields a precise mapping of the
underlying functions, free from both inter-subject and inter-site variabil-
ity. Additionally, the dataset comes with high-resolution anatomical and
diffusion images, to achieve a fine anatomical characterization of these
brains.

(*) Actually, the dataset comprises data from 13 participants, but
one, sub-02 dropped out. It is included in the datasets for
completeness but only has partial acquisitions.

Dataset content overview
========================

This is an overview of the dataset content.

More information and updates are available at http://project.inria.fr/IBC/

The data in all IBC release are organized as follows:

./task*.json: Descritpion of the tasks used in this release
./participants.tsv Description of the participants
./sub-XX subject directories
./sub-XX/ses-YY session directories

Note that there are many sessions per subject and that the session are
numbered according to acquisition date, which yields different actual
acquisitions depending on the subjects. The data are thus much better
by the MRI file names, than by session id.

Session directories comprise the following subdirectories:
* anat/ for T1w-, T2w- and flair image
* dwi/ for diffusion-weighted data
* fmap/ for field maps
* func/ for functional data

Note that diffusion-weighted, T2-weighted and FLAIR images are not
meant to be used for advanced neuroimaging investigations. They were
acquired to check the absence of conspicuos abnormalities at screening
stage. High-resolution anatomical and diffusion-weighted images will
be procided in the future.

Regarding functional data, the file naming conventions is
sub-XX_ses-YY_task-ZZZ_acq-AA_bold.nii.gz,
where
* XX is the subject id
* YY is the session id
* ZZZ is one of the twelve tasks used
* 'AA' is either ap pa, corresponding to the phase encoding direction

These fMRI files are provided together with
sub-XX_ses-YY_task-ZZZ_acq-AA_events.tsv files,
that describe the corresponding events.

In the same folder, the 'ap' and 'pa' variant of a given acquisition
are always found. the corresponding fiel maps are available in the
fmap folder to correct distortions. test-retest or fixed effectsmodels
can be computed across the corresponding data after suitable
preprocessing.

More information and material
=============================

The functional protocols used can be dowloaded at:
https://github.com/hbp-brain-charting/public_protocols

Analysis scripts can be found at:
https://github.com/hbp-brain-charting/public_analysis_code

Contact
=======
Bertrand Thirion, [e-mail]