Skip to content
Open Imaging Index
MRI PET Brain

MEMENTO

MEMENTO: cohort of outpatients from French research memory centres (deterMinants and Evolution of AlzheiMer's disEase aNd relaTed disOrders)

Brain MRI, optional FDG PET, biobank and clinical data of memory-clinic patients without dementia, with subjective complaints or mild cognitive impairment, followed for 5 years to study how Alzheimer's disease and related disorders develop. 2,323 adults in France. Access by approved proposal.

Go to dataset Edit Report

Overview

MEMENTO is a French clinic-based cohort set up under the 2008-2012 national Alzheimer plan to follow people who consult a memory clinic before they develop dementia. It is run from the clinical epidemiology unit (CIC 1401-EC) in Bordeaux, sponsored by Bordeaux University Hospital and registered as NCT01926249. Imaging was harmonised across sites by CATI. The SHIVA-CMB microbleed detector (Tsuchida et al. 2024) was evaluated on its baseline T2*-weighted scans.

Composition

2,323 non-demented adults were enrolled at 26 university-based memory clinics between April 2011 and June 2014. At baseline 61.8% were women and the mean age was 70.9 years (SD 8.7). The design paper sorts 2,313 of them into five cognitive groups: 370 with an isolated cognitive complaint and 1,943 with single- or multi-domain, amnestic or non-amnestic mild cognitive impairment. Participants were seen every six months for five years, with clinical, neuropsychological, lifestyle and biobank data; the DPUK record lists 1,481 participants at the fifth annual visit.

Acquisition

The harmonised MRI protocol is compatible with ADNI and contains a 3D T1-weighted scan, 2D FLAIR, a 2D T2*-weighted gradient echo, a 2D T2-weighted turbo spin echo and, as options, resting-state BOLD and diffusion. 86% of participants were scanned at 3 T and the rest at 1.5 T, on systems from several vendors. Structural MRI was planned at baseline, month 24 and month 48. FDG PET was optional and done in 60% of participants at baseline; the DPUK record also lists amyloid PET. In SHIVA-CMB, the T2*-GRE used TR 650 ms, TE 20 ms, flip angle 20 degrees and 1.0 x 1.0 x 2.5 mm voxels.

Annotations

No voxel labels are described. For 1,992 participants, a trained rater counted cerebral microbleeds on the baseline T2*-GRE with the Microbleed Anatomical Rating Scale, checked by a neuroradiologist; SHIVA-CMB reports 1,503 microbleeds in total. Hippocampal volume, white matter lesion volume, cortical thickness and regional FDG uptake were derived centrally.

Known limitations

  • Not open. Each project needs an approved proposal, and the charter requires specific authorisation for raw images.
  • Analysis through DPUK happens on a remote portal; whether raw images can be released outside it is not stated.
  • Sex, field strength and PET counts are published as percentages only, so they are not in the stats table.
  • The number of participants with a T2*-GRE is known only for the 1,992 used in SHIVA-CMB.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 2,323 subjects.

Age

mean 70.9 ± 8.7

No age bins reported.

Table 3 in Dufouil et al. 2017, MEMENTO study design and baseline findings, Alzheimer's Research & Therapy (PMC5576287)

Condition

Groups can overlap

  • Mild cognitive impairment 1,943 84%

Table 4 in Dufouil et al. 2017, MEMENTO study design and baseline findings, Alzheimer's Research & Therapy (PMC5576287)

Contrast / sequence

Groups can overlap

  • T2*-weighted 1,992 86%

Table 1 in Tsuchida et al. 2024, SHIVA-CMB, Scientific Reports (PMC11680838)

Country

  • France 2,323 100%

Methods, Study design in Dufouil et al. 2017, MEMENTO study design and baseline findings, Alzheimer's Research & Therapy (PMC5576287)

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
Application

A research proposal is reviewed and approved

Access page

Access to Memento data and Publication Charter

Every use of MEMENTO data is a proposal that the MEMENTO principal investigators and scientific coordination committee must approve. Raw MRI and PET images need specific authorisation, a contract may be required, and every abstract and manuscript must be approved by the committee before submission and list the MEMENTO Study Group.

Proposals go to the MEMENTO secretariat with the completed charter, or are filed online through the Dementias Platform UK (DPUK), where analysis runs remotely on the DPUK portal and only aggregate results leave it under DPUK and MEMENTO control. The DPUK route also falls under DPUK's own access terms, which this entry does not break down. The charter says access to raw MRI and PET images held at CATI needs specific authorisation.

Original license text Version read: v4.0, May 1st 2019 Checked 2026-10-08

What you can do

  • Not stated
  • Not stated
  • Train ML models Not stated
  • Create derived data Not stated
  • Conditional

What you can share

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

What you must do

  • Yes
  • Share alike No
  • Conditional
  • Conditional
  • Yes
  • Release code No
  • No
  • Delete after use No

Limits

  • No
  • No

Citation

Dufouil C, Dubois B, Vellas B, et al. Cognitive and imaging markers in non-demented subjects attending a memory clinic: study design and baseline findings of the MEMENTO cohort. Alzheimer's Research & Therapy 9, 67 (2017). doi:10.1186/s13195-017-0288-0

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.