Skip to content
Open Imaging Index
MRI Brain

Robert Debré Hospital autism MRI

Robert Debré Hospital (Paris) autism MRI cohort

Structural and resting-state brain MRI of people with autism and controls from a Paris hospital, used as the unseen test site of a challenge to predict autism diagnosis from MRI. 247 people, 49 with autism. Not released.

Go to dataset Edit Report

Overview

The Robert Debré Hospital autism MRI cohort is a set of brain scans collected at Robert Debré Hospital (AP-HP) in Paris by its child and adolescent psychiatry department together with the Institut Pasteur. It has not been released. Its only description as a whole comes from Traut et al. (2022), who used it as an unseen external site in the IMPAC challenge, where teams built models to predict an autism diagnosis from MRI. The other challenge data came from ABIDE I and II.

Composition

The cohort holds 247 people: 49 with autism spectrum disorder and 198 controls. The groups differ a lot in age and sex. Autistic participants were mostly children and adolescents (mean age 14.2, range 4.7 to 43.7, 39 male and 10 female), while controls were mostly adults (mean 32.4, range 4.0 to 70.8, 85 male and 113 female). Recruitment was family-based: 107 participants are probands and 140 are relatives, and 134 of the controls are relatives. Cognitive level was rated by clinical judgement backed by IQ tests and was normal in 227 people, borderline in 10 and delayed in 9. 56 of the 247 subjects are also part of ABIDE II.

Acquisition

All scans were acquired on a 1.5 T scanner, unlike the 3 T scanners of the other challenge sites. The paper does not name the scanner vendor or give sequence parameters. The challenge data, this site included, consisted of anatomical MRI and resting-state functional MRI, processed with FreeSurfer, FSL and AFNI. Diagnoses were supported by the ADI and ADOS instruments alongside clinical assessment.

Annotations

No image annotations. Labels are diagnosis (autism or control), age, sex and family status.

Known limitations

  • Not shared. The images stay with the holders, the paper's data availability statement covers only the public ABIDE part, and no access process or terms are published.
  • Age and sex are strongly confounded with diagnosis, and many controls are relatives of autistic probands.
  • The paper gives no per-subject breakdown of which contrasts are available.
  • 56 subjects overlap with the ABIDE II IP_1 collection, so pooled analyses must remove duplicates.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 247 subjects.

Sex

  • Female 123 50%
  • Male 124 50%

Supplemental Table 1 in Traut et al. 2022, medRxiv supplemental material

Condition

Groups can overlap

  • Healthy control 198 80%
    Age mean 32.4 ± 14.7, median 34.4, range 4 to 70.8
  • Autism spectrum disorder 49 20%
    Age mean 14.2 ± 8.1, median 13, range 4.7 to 43.7

Supplemental Table 1 in Traut et al. 2022, medRxiv supplemental material

Country

  • France 247 100%

p. 4 in Traut et al. 2022, medRxiv preprint v3 (posted 31 March 2022) of the NeuroImage paper

Age

mean 28.8 ± 15.4, median 27.9, range 4 to 70.8

No age bins reported.

Supplemental Table 1 in Traut et al. 2022, medRxiv supplemental material

Modality

  • MRI 247 100%

p. 4 in Traut et al. 2022, medRxiv preprint v3 (posted 31 March 2022) of the NeuroImage paper

Field strength

  • 1.5 T 247 100%

p. 4 in Traut et al. 2022, medRxiv preprint v3 (posted 31 March 2022) of the NeuroImage paper

Condition by sex

Reported cross table. A dot marks a cell the source does not give.

FemaleMale
Healthy control11385
Autism spectrum disorder1039

Supplemental Table 1 in Traut et al. 2022, medRxiv supplemental material

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

Access
Not shared

The data are not available outside the institution or study that holds them

Traut et al. (2022) call the Robert Debré data an unpublished dataset and kept it in the private test set of the challenge, which participants never saw. Their data availability statement covers only the public ABIDE subset, the code and the preprocessing scripts. 56 of the 247 subjects were also released as the IP_1 collection of ABIDE II.

Access page

No published license or data use terms

The data holder has published no license and no data use terms. Any use, sharing or commercial right has to be agreed with the data holder, and nothing can be assumed to be allowed. Default copyright and data protection law still apply.

Neither the paper, its supplement nor the challenge repository (github.com/neuroanatomy/autism-challenge) states a license or data use terms for the Robert Debré images.

Original license text Checked 2026-10-08

What you can do

  • Commercial use Not stated
  • Not stated
  • Train ML models Not stated
  • Create derived data Not stated
  • Publish results Not stated

What you can share

  • Share the data Not stated
  • Share derived data Not stated
  • Share trained models Not stated

What you must do

  • Cite or credit Not stated
  • Share alike Not stated
  • Sign an agreement Not stated
  • Ethics approval Not stated
  • Manuscript review Not stated
  • Release code Not stated
  • Return results Not stated
  • Delete after use Not stated

Limits

  • No re-identification Not stated
  • Location limits Not stated

Citation

Traut N, Heuer K, Lemaître G, et al. Insights from an autism imaging biomarker challenge: promises and threats to biomarker discovery. NeuroImage 255:119171 (2022). doi:10.1016/j.neuroimage.2022.119171

Sources

Every number on this page comes from one of these documents. Each chart names the table or page it is taken from. The raw numbers are in stats.csv.