Skip to content

3T Diffusion MRI Inter-Site Reproducibility Dataset for Cerebral Small Vessel Disease

Latest
Latest

Choose a tag to compare

View all tags
@data-hcp data-hcp released this 25 Mar 02:11
lmu-svd
b83e013
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

CADASIL patients scanned on Siemens Prisma and Skyra within 24 hours
data-others/intersite-svd

This dataset provides inter-site / inter-scanner diffusion MRI from 10 patients with CADASIL, a genetically defined cerebral small vessel disease (SVD). Each participant was scanned on two different 3 Tesla Siemens MRI scanners — Magnetom Prisma and Magnetom Skyra — within less than 24 hours.

The dataset was created to evaluate the reproducibility of diffusion-derived metrics across scanners and is intended to support methodological development, clinical translation, and multi-center trial preparation in cerebral small vessel disease.

Dataset Overview

  • Title: 3T diffusion MRI inter-site reproducibility dataset for cerebral small vessel disease
  • Alternative titles:
    • Intersite-SVD
    • Interscanner-SVD
  • Subjects: 10 CADASIL patients
  • Condition: Cerebral small vessel disease (CADASIL)
  • Design: Same subjects scanned on 2 scanners within <24 hours
  • Data type: Diffusion MRI only
  • Format: NIfTI, organized in BIDS format
  • Archive: intersite-svd.zip
  • Size: 11.1 GB

Terms of Use

This dataset is provided under the:

Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

It may be freely downloaded and used for non-commercial purposes. Use of the dataset should acknowledge the associated Neurology paper.

Scientific Motivation

Diffusion MRI metrics are widely used as sensitive markers of microstructural tissue damage in cerebral small vessel disease. For these measures to be useful in clinical routine and multi-center trials, they need to be reproducible across scanners and sites.

This dataset addresses that need by providing a controlled within-subject comparison across two 3T scanners located in separate buildings with independent infrastructure at the same hospital.

The dataset has already been used in:

Konieczny & Dewenter et al., Neurology (2020)
https://doi.org/10.1212/WNL.0000000000011213

Participants

Participants were recruited at the Institute for Stroke and Dementia Research, Munich, Germany.

Cohort summary

  • n = 10
  • Diagnosis: CADASIL, confirmed by molecular genetic testing or skin biopsy

Clinical characteristics

Characteristic Value
Age, median (IQR) 55.5 (13.3) years
Female 5 (50%)
Hypertension 3 (30%)
Hypercholesterolaemia 5 (50%)
Diabetes 0 (0%)
Current or past smoking 3 (30%)
WMH volume [% ICV], median (IQR) 7.67 (2.62)
Lacune count, median (IQR) 3 (7.75)
Lacune volume [% ICV], median (IQR) 0.01 (0.02)
Microbleed count, median (IQR) 0.5 (1.75)
Brain volume [% ICV], median (IQR) 75.7 (7.03)

Abbreviations:
ICV = intracranial volume
IQR = interquartile range
WMH = white matter hyperintensities

MRI Acquisition

Participants were scanned on two Siemens 3T systems at LMU University Hospital:

  • Siemens Magnetom Skyra
  • Siemens Magnetom Prisma

The scanners were in separate buildings, which is why the dataset is treated as inter-site despite being from the same hospital.

Key protocol parameters were harmonized across scanners, including:

  • b-values
  • number of diffusion directions
  • spatial resolution

Because the Skyra has a weaker gradient system, it required longer TE and TR.

MRI protocol comparison

Parameter Magnetom Skyra Magnetom Prisma
Coil channels 64 (head-neck) 64 (head-neck)
TR [ms] 3800 3200
TE [ms] 104.8 74
Flip angle [°] 90 90
In-plane resolution [mm] 2 × 2 2 × 2
Slice thickness [mm] 2 2
Base resolution 120 120
Number of slices 75 75
b-values [s/mm²] 1000 / 2000 1000 / 2000
Directions per b-value 30 / 60 30 / 60
b=0 images 10 10
Receiver bandwidth [Hz/px] 1894 1954
Parallel imaging factor 2 2
Multi-band factor 3 3

Data Organization

The dataset is stored in NIfTI format and follows the BIDS standard.
bval and bvec files follow FSL convention.

Included data

  • Raw diffusion MRI
    • converted from DICOM using dcm2niix
  • Preprocessed diffusion MRI
    • processed as described in the manuscript

Directory structure

/sub-01
    /ses-prisma
        /dwi_raw
        /dwi_preprocessed
    /ses-skyra
        /dwi_raw
        /dwi_preprocessed
Assets 63
Loading