Skip to content
Open Imaging Index
MRI Brain

BrainLat

The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds

Brain MRI, resting-state EEG and cognitive tests from Latin American patients with Alzheimer's disease, frontotemporal dementia, Parkinson's disease or multiple sclerosis and healthy controls, shared to develop affordable biomarkers for underrepresented populations. 780 people from five countries.

Go to dataset Edit Report

Overview

BrainLat is a multimodal collection of brain imaging, resting-state EEG, cognitive scores and demographics from people with neurodegenerative diseases and healthy controls in Latin America. It was assembled by the Latin American Brain Health Institute with clinical sites of the ReDLat consortium, which harmonised recruitment and neuropsychological assessment across centres. The aim is to give researchers data from populations that are rare in other dementia cohorts and to support low-cost, scalable biomarkers. The data are hosted on Synapse and described by Prado et al. in Scientific Data (2023).

Composition

The paper reports 780 participants: 278 with Alzheimer's disease, 163 with behavioural variant frontotemporal dementia, 57 with Parkinson's disease, 32 with multiple sclerosis and 250 healthy controls. 453 are women. Mean age is 62.7 years (range 21 to 89); the multiple sclerosis group is much younger than the others. Participants came from eight sites in Argentina, Chile, Colombia, Mexico and Peru. Some Alzheimer's patients from one Colombian site carry PSEN1 mutations. Not every participant has every modality: according to the per-group table, about 710 have anatomical T1-weighted MRI, about 440 resting-state fMRI and about 360 diffusion MRI, and a smaller group has only EEG. Cognitive (MoCA, IFS and others) and disease-specific clinical scales are supplied as CSV files.

Acquisition

MRI was acquired on 1.5 T and 3 T scanners from GE, Siemens and Philips within six months of the clinical visit. The protocol includes a T1-weighted MPRAGE, multi-echo resting-state BOLD with eyes open and diffusion imaging; the paper also mentions T2-FLAIR but gives no counts for it. EEG was recorded for ten minutes, eyes closed, on a 128-channel Biosemi system at every site and lightly preprocessed. Images were defaced with PyDeface and organised in BIDS, with MRIQC quality metrics in the derivatives.

Annotations

No image annotations. Labels are the clinical diagnosis, sex, age, education, handedness, cognitive test scores and disease severity scales.

Known limitations

  • Imaging was not harmonised across sites and scanners.
  • Several tables in the paper are internally inconsistent: the control rows of the imaging table add up to more than the 250 controls, and some per-site sex ratios do not match their counts.
  • Disease duration, genetics and socioeconomic data are not included.
  • MRI download needs a Synapse account and a short request form; EEG is in an open folder of the same project.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 780 subjects.

Sex

  • Female 453 58%
  • Male 327 42%

Table 2 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Contrast / sequence

Groups can overlap

  • T1-weighted 712 91%
  • BOLD fMRI 441 57%
  • Diffusion-weighted 363 47%

Table 4 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Field strength

  • 3 T 451 58%
  • 1.5 T 265 34%

Table 4 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Country

  • Chile 291 37%
  • Argentina 204 26%
  • Colombia 175 22%
  • Peru 81 10%
  • Mexico 29 4%

Table 3 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Condition by field strength

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

3 T1.5 T
Healthy control17781
Alzheimer's disease91156
Frontotemporal dementia9728
Parkinson's disease55·
Multiple sclerosis31·

Table 4 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Age

mean 62.7 ± 9.5, range 21 to 89

No age bins reported.

Abstract; Table 2, Abstract in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Modality

  • MRI 716 92%

Table 4 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Condition

Groups can overlap

  • Alzheimer's disease 278 36%
    Age mean 72.2 ± 7.9
  • Healthy control 250 32%
    Age mean 67.9 ± 8.9
  • Frontotemporal dementia 163 21%
    Age mean 65.1 ± 10.5
  • Parkinson's disease 57 7%
    Age mean 69.9 ± 11.2
  • Multiple sclerosis 32 4%
    Age mean 38.5 ± 8.9

Table 2 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Condition by sex

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

FemaleMale
Alzheimer's disease167111
Healthy control16486
Frontotemporal dementia7786
Parkinson's disease1938
Multiple sclerosis266

Table 2 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

Condition by contrast / sequence

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

T1-weightedBOLD fMRIDiffusion-weighted
Healthy control·198188
Alzheimer's disease2458654
Frontotemporal dementia1249791
Parkinson's disease553030
Multiple sclerosis3130·

Table 4 in Prado et al. 2023, The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds, Scientific Data

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
Free registration

Create an account and accept terms online

The Synapse project page says the EEG data can be taken directly from its data folder. For the MRI data, users with a Synapse account fill in a short Google form (name, institution, e-mail and Synapse username) or e-mail the named contact at Universidad de San Andrés.

Access page

Creative Commons Zero 1.0 Universal

Public domain dedication. Do anything with the data, including commercial use, without asking and without having to give credit.

The Synapse project wiki states that the BrainLat Dataset, © 2023 by Pavel Prado, Vicente Medel and Agustín Ibañez, is licensed under CC0.

Original license text Version read: 1.0 Checked 2026-10-07

What you can do

  • Yes
  • Yes
  • Yes
  • Yes
  • Yes

What you can share

  • Yes
  • Yes
  • Yes

What you must do

  • No
  • Share alike No
  • No
  • No
  • Manuscript review No
  • Release code No
  • Return results No
  • Delete after use No

Limits

  • No
  • Location limits No

Citation

Prado P, Medel V, Gonzalez-Gomez R, Sainz-Ballesteros A, Vidal V, Santamaría-García H, Moguilner S, Mejia J, Slachevsky A, Behrens MI, Aguillon D, Lopera F, Parra MA, Matallana D, Maito MA, Garcia AM, Custodio N, Funes AA, Piña-Escudero S, Birba A, Fittipaldi S, Legaz A, Ibañez A. The BrainLat project, a multimodal neuroimaging dataset of neurodegeneration from underrepresented backgrounds. Scientific Data 10, 889 (2023). doi:10.1038/s41597-023-02806-8

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.