MRi-Share
MRi-Share: brain MRI substudy of the i-Share student health cohort
Multimodal 3 T brain MRI of 1,832 healthy university students aged 18 to 35 in Bordeaux, France, scanned 2015 to 2017 with a UK Biobank-like protocol: T1w, FLAIR, SWI, multi-shell diffusion and 15 min resting-state fMRI. Access by approved i-Share collaboration.
Overview
MRi-Share is the brain imaging arm of i-Share, a French cohort study of student health coordinated from the University of Bordeaux. The Neurofunctional Imaging Group (GIN-IMN) scanned students from the Bordeaux site to describe how brain structure and function change in the late teens and twenties, to look for early markers of later-life disease, and to serve as an imaging data platform for other researchers. The protocol copies the UK Biobank brain imaging protocol closely so that the two cohorts can be analysed together across the adult age range.
Composition
Of 2,000 students who met the study physicians between October 2015 and June 2017, 1,870 were scanned. After one withdrawal, one scanner failure and 36 exclusions for incidental findings that needed medical referral, the described sample holds 1,832 participants: 1,320 women and 512 men, mean age 22.1 years (range 18.1 to 34.9). 1,722 were younger than 26. Participants were healthy volunteers; exclusion criteria were age over 35, pregnancy or nursing, claustrophobia and MRI contraindications.
Acquisition
All scans come from one Siemens Prisma 3 T scanner with a 64-channel head coil, acquired between November 2015 and November 2017 in sessions of about 45 minutes. Each session has a 3D MPRAGE T1w and a 3D SPACE FLAIR at 1 mm isotropic, an axial SWI (0.8 x 0.8 x 3.0 mm), multi-shell diffusion at 1.75 mm with b-values 300, 1000 and 2000 s/mm2, and a resting-state fMRI run of about 15 minutes (multiband 6, TR 850 ms, 2.4 mm). All 1,832 have T1w and FLAIR; SWI, diffusion and resting-state fMRI are available for 1,830, 1,823 and 1,814. T1w and FLAIR images are defaced with the UK Biobank face mask.
Annotations
No manual labels are described. The GIN processing platform (ABACI) produces image-derived phenotypes such as cortical thickness, surface area, tissue volumes, diffusion tensor and NODDI metrics, and resting-state connectivity. Global phenotypes for 1,831 participants are openly available as a table; the images are not.
Known limitations
- Not public. Access needs an approved i-Share collaboration proposal, a signed charter and a contribution to costs.
- About 72% of participants are women, which reflects the i-Share cohort, and almost all are under 26.
- Single site and single scanner, so it does not test generalisation across hardware.
- Cross-sectional: each participant was scanned once.
Cohort
Aggregate numbers from the sources below. Bars are relative to the 1,832 subjects.
Sex
- Female 1,320 72%
- Male 512 28%
Age
mean 22.1± 2.3 · range 18.1 to 34.9Age by sex
Reported cross table. Missing cells were not published (fewer than 10 subjects or not reported).
- 1,25218-25470
Contrast combinations
How many subjects have exactly each set of contrasts.
| T1w | FLAIR | Subjects with exactly this set |
|---|---|---|
2 |
Contrast / sequence
subjects, values can overlap
- T1-weighted 1,832 100%
- FLAIR 1,832 100%
- Susceptibility-weighted 1,830 100%
- Diffusion-weighted 1,823 100%
- BOLD fMRI 1,814 99%
Scanner vendor
subjects
- Siemens Healthineers 1,832 100%
Field strength
subjects
- 3 T 1,832 100%
Country
subjects
- France 1,832 100%
License and access
Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer
A research proposal is reviewed and approved
Ancillary Studies Charter, University of Bordeaux / i-Share study
Every use of i-Share data, including the MRi-Share imaging platform, is an ancillary study that the i-Share executive committee must approve. The lead researcher signs the charter, pays for access, may use the data only for the declared analyses and by the named people, destroys the data at the end, and follows the publication rules, which make the MRi platform lead a co-signatory of every article that uses its data.
The data descriptor says the images cannot go to a public repository under French rules and that requests go to the i-Share Scientific Collaborations Coordinator with a letter of intent, followed by an application package reviewed by the PI, the Steering Committee and operational staff. The i-Share site asks applicants to sign the i-Share ancillary studies charter, and the charter makes requesters fund the cost of access. Global image-derived phenotypes of 1,831 participants (age, sex and 15 summary measures, no images) are open on Dryad under CC0 and are not covered by this entry.
What you can do
- Not stated
- Train ML models Not stated
- Not stated
- Conditional
What you can share
- No
- Share derived data Not stated
- Share trained models Not stated
What you must do
- Yes
- Share alike No
- Yes
- Conditional
- Conditional
- Conditional
- Conditional
- Yes
Limits
- Conditional
- No
Citation
Tsuchida A, Laurent A, Crivello F, et al. The MRi-Share database: brain imaging in a cross-sectional cohort of 1870 university students. Brain Structure and Function 226, 2057-2085 (2021). doi:10.1007/s00429-021-02334-4
All numbers
Every number on this page, as stored in stats.csv, with its source.
| Measure | Breakdown | Value | Source |
|---|---|---|---|
| Subjects | total final sample: 1,870 were scanned, 1 withdrew, 1 scan failed and 36 with incidental findings needing referral were excluded | 1,832 | tsuchida2020 Table 1 |
| Subjects | sex=female | 1,320 | tsuchida2020 Table 1 |
| Subjects | sex=male | 512 | tsuchida2020 Table 1 |
| Subjects | age=18-25 under 26 years old | 1,722 | tsuchida2020 Table 1 |
| Subjects | sex=female;age=18-25 under 26 years old | 1,252 | tsuchida2020 Table 1 |
| Subjects | sex=male;age=18-25 under 26 years old | 470 | tsuchida2020 Table 1 |
| Mean age | total | 22.1 | tsuchida2020 Table 1 |
| Age SD | total | 2.3 | tsuchida2020 Table 1 |
| Minimum age | total | 18.1 | tsuchida2020 Table 1 |
| Maximum age | total | 34.9 | tsuchida2020 Table 1 |
| Subjects | contrast=T1w 3D MPRAGE 1.0 mm isotropic | 1,832 | tsuchida2020 Table 2 |
| Subjects | contrast=FLAIR 3D SPACE FLAIR 1.0 mm isotropic | 1,832 | tsuchida2020 Table 2 |
| Subjects | contrast=swi 2D axial 0.8 x 0.8 x 3.0 mm | 1,830 | tsuchida2020 Table 2 |
| Subjects | contrast=dwi multi-shell, b = 300, 1000 and 2000 s/mm2 | 1,823 | tsuchida2020 Table 2 |
| Subjects | contrast=bold resting-state fMRI, ~15 min; excludes 17 missing or incomplete runs and 1 run with a wrong TR | 1,814 | tsuchida2020 Table 2 |
| Subjects | contrast_set=FLAIR+T1w SWI, DWI and rs-fMRI missing | 2 | tsuchida2020 Data Acquisition, MRI acquisition |
| Subjects | vendor=siemens one Siemens 3T Prisma with a 64-channel head coil | 1,832 | tsuchida2020 Data Acquisition, MRI acquisition |
| Subjects | field_strength=3 | 1,832 | tsuchida2020 Data Acquisition, MRI acquisition |
| Subjects | country=FR Bordeaux | 1,832 | tsuchida2020 Data Acquisition, MRI acquisition |
Sources
The keys used in the table above.
- tsuchida2020 Tsuchida et al. 2020, The MRi-Share database, bioRxiv preprint v1 paper
- tsuchida2021 Tsuchida et al. 2021, The MRi-Share database, Brain Structure and Function paper
- gin-mri-share GIN-IMN MRi-Share project page website
- i-share-procedures i-Share research site, complete process and procedures for collaboration website
- dryad-mri-share-idps Dryad: Data from The MRi-Share database, global imaging-derived phenotypes (CC0) data