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

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

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

Fast Kurtosis Imaging (Sequence 139) with T1-weighted MRI

This dataset provides diffusion kurtosis MRI (DKI) acquired using a fast kurtosis sequence (the 139 sequence) together with T1-weighted structural MRI from 44 patients with mild cognitive impairment (MCI) and 23 cognitively normal controls. The T1-weighted images have been skull-stripped for anonymity.

The dataset is suitable for studies of microstructural alterations in early cognitive decline and for development or benchmarking of diffusion kurtosis processing pipelines.

Dataset Overview

• Title: Diffusion kurtosis MRI dataset
• DOI: https://doi.org/10.6084/m9.figshare.30850301
• Release date: 2025-12-10
• Author: Simon Eskildsen
• Total size: ~385.84 MB

Participants

• MCI patients: 44
• Cognitively normal controls: 23
• Total subjects: 67

Modalities

• Diffusion kurtosis MRI (DKI)
• T1-weighted MRI (skull-stripped)

Data Organization

• DKI images:
  Provided as separate NIfTI files, one file per diffusion direction.

• b-values and b-vectors:
  Stored as text files in the same folders as the diffusion images.

• Structural MRI:
  T1-weighted images are included and have been skull-stripped to protect subject anonymity.

• Demographics:
  Subject-level demographic information, including group identifiers, is available in demographics.csv.

This organization makes the dataset straightforward to use in custom diffusion pipelines or for conversion into other diffusion MRI formats.

Scientific Context

Please refer to the following paper for a detailed description of the study population and scientific background:

Nielsen RB, Parbo P, Ismail R, Dalby RB,
Tietze A, Brændgaard H, Gottrup H,
Brooks DJ, Østergaard L and Eskildsen SF (2025)
Diffusion kurtosis imaging detects cortical microstructural alterations in amyloid-positive MCI patients.
Front. Dement. 4:1725754.
doi: 10.3389/frdem.2025.1725754

Article link:
https://www.frontiersin.org/journals/dementia/articles/10.3389/frdem.2025.1725754/abstract

Suggested Uses

This dataset can be used for:

• Diffusion kurtosis model fitting
• Comparison of DKI and conventional DTI metrics
• Studies of cortical microstructural alterations in MCI
• Biomarker development for early cognitive impairment
• Method development and validation for compact clinical diffusion datasets
• Multimodal analyses combining DKI and structural MRI

License

This dataset is distributed under the:

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License (CC BY-NC-SA 4.0)

• Attribution required
• Non-commercial use only
• Share alike: derivative works must use the same license

Funding

• DFF-4004-00305

Citation

Please cite both the dataset and the associated publication:

Dataset:
Eskildsen, S.
Diffusion kurtosis MRI dataset
https://doi.org/10.6084/m9.figshare.30850301

Associated paper:
Nielsen RB, Parbo P, Ismail R, Dalby RB, Tietze A, Brændgaard H, Gottrup H, Brooks DJ, Østergaard L and Eskildsen SF (2025)
Diffusion kurtosis imaging detects cortical microstructural alterations in amyloid-positive MCI patients.
Front. Dement. 4:1725754.
https://doi.org/10.3389/frdem.2025.1725754

Notes for This Repository

The data-others/dki-mci release may include:

• Harmonized file naming
• Preorganized diffusion inputs for downstream processing
• Optional DSI Studio derivatives (*.sz, *.fz)
• QC summaries where available

Users should consult the original dataset and publication for full methodological details and cohort description.

This dataset contains structural and diffusion MRI data together with derived FA, TBSS, and VBM results for patients diagnosed with Alzheimer’s Disease (AD), Bipolar Disorder (BD), and healthy controls (HC).
It includes corresponding clinical and biomarker information organized in structured .csv files, supporting both neuroimaging and clinical correlation studies.

The dataset has been used in multiple peer-reviewed publications investigating white matter integrity, biomarker correlation, and multivariate diagnostic modeling of neurodegenerative and psychiatric disorders.

License

Creative Commons Attribution 4.0 International (CC BY 4.0)

Citation

Besga, A., Graña, M., & Chyzhyk, D. (2020).
Alzheimer’s Disease versus Bipolar Disorder versus Healthy Control MRI Data and Processed Results [Data set]. Zenodo.
https://doi.org/10.5281/zenodo.3935636

Source

https://doi.org/10.5281/zenodo.3935636
Contact: ariadna.besga@ehu.eus
Institutions: University of the Basque Country (UPV/EHU), University of Navarra, University of the Basque Country
Funding: European Commission — CybSPEED: Cyber-Physical Systems for PEdagogical Rehabilitation in Special Education (Grant No. 777720)

Dataset Information

Clinical Data

The Clinical_data folder contains .csv files readable in Python, R, MATLAB, or any spreadsheet software.
Each file corresponds to a specific biomarker.

Key files include:

• clinical_data_id_age_gender.csv: anonymized ID, diagnostic key, age, and gender.
• clinical_data_corrected.csv: deprecated, can be ignored.

Diagnostic key:

• crl = Healthy control
• tb = Bipolar disorder
• ea = Alzheimer’s disease

Publications Using This Dataset

1. Graña et al. (2011) — Computer Aided Diagnosis system for Alzheimer’s Disease using DTI features, Neuroscience Letters, 502(3):225–229.
2. Besga et al. (2012) — Discovering Alzheimer’s disease and bipolar disorder white matter effects using DTI, Neuroscience Letters, 520(1):71–76.
3. Termenon et al. (2013) — Lattice ICA feature selection on DWI for Alzheimer’s classification, Neurocomputing, 114:132–141.
4. Besga et al. (2015) — Discrimination between AD and late-onset BD using multivariate analysis, Frontiers in Aging Neuroscience, 7:231.
5. Besga-Basterra et al. (2016) — Eigenanatomy on FA imaging differentiates AD and BD, Current Alzheimer Research, 13(5):557–565.
6. Besga et al. (2017) — White matter tract integrity and inflammation in AD vs BD, Frontiers in Aging Neuroscience, 9:179.

Purpose

This dataset supports research in neurodegenerative and affective disorder differentiation, enabling:

• Quantitative analysis of white matter microstructure (FA, TBSS).
• Morphometric evaluation via SPM-based VBM.
• Correlation between MRI-derived features and clinical biomarkers.
• Machine learning model training for computer-aided diagnosis (CAD) of AD vs BD.

Keywords

MRI • Diffusion MRI • FA • TBSS • VBM • Alzheimer’s Disease • Bipolar Disorder • Healthy Control • FSL • SPM • Neurodegeneration • White Matter Integrity • Biomarkers

A curated dataset and analysis pipelines for a cross‑sectional Parkinson’s disease (PD) study, designed to support reproducible research in diffusion MRI.

📄 License

This work is licensed under Creative Commons Attribution‑ShareAlike (CC BY‑SA).

• NITRC Image Repository: parktdi

📋 Overview

This project contains data and analysis pipelines for a set of 53 subjects in a cross‑sectional PD study:

• 27 PD patients
• 26 age, sex, and education‑matched controls

The diffusion‑weighted images (DWI) were acquired with:

• 120 unique gradient directions
• b = 1000 and b = 2500 s/mm²
• Isotropic 2.4 mm³ voxels
• Twice‑refocused spin‑echo sequence to minimize eddy‑current distortions

This dataset provides multi-site diffusion MRI data from 10 patients diagnosed with the genetically defined small vessel disease CADASIL, scanned on two different 3 T Siemens MRI scanners (Magnetom Prisma and Magnetom Skyra) within 24 hours.
It enables evaluation of inter-scanner reproducibility of diffusion metrics in cerebral small vessel disease (SVD) and supports harmonization studies for multi-center neuroimaging trials.

The dataset was used in the publication:
Konieczny & Dewenter et al., Neurology, 2020 — https://doi.org/10.1212/WNL.0000000000011213

License

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

Citation

Dewenter, A., Gesierich, B., & Duering, M. (2020).
3T diffusion MRI inter-site reproducibility dataset for cerebral small vessel disease [Data set]. Zenodo.
https://doi.org/10.5281/zenodo.16925507

Source

https://doi.org/10.5281/zenodo.16925507
Contact: marco.duering@med.uni-muenchen.de
Institution: Institute for Stroke and Dementia Research (ISD), LMU Munich, Germany
Associated publication: Neurology, 96(5), e698–e708 (2020)

Dataset Information

MRI Acquisition Protocol

Both scanners were located in separate hospital buildings with independent infrastructure, establishing a true inter-site condition despite institutional proximity.

Data Structure and Preprocessing

The dataset follows the BIDS organization scheme:

/sub-01
   /ses-prisma
       /dwi_raw
       /dwi_preprocessed
   /ses-skyra
       /dwi_raw
       /dwi_preprocessed

• Raw data converted from DICOM via dcm2niix
• Preprocessed data includes denoising (MRtrix dwidenoise), Gibbs correction (mrdegibbs), and motion/eddy distortion correction (FSL TOPUP, EDDY).

Purpose

This dataset serves as a benchmark for assessing inter-scanner reproducibility and harmonization of diffusion MRI measures across 3 T sites.
It is particularly relevant for researchers working on:

• Cross-site calibration of diffusion metrics in small vessel disease
• Methodological validation of preprocessing and harmonization pipelines
• Clinical reproducibility of DTI/DKI biomarkers in multicenter trials

Keywords

Diffusion MRI • DTI • CADASIL • Small Vessel Disease • Inter-site Reproducibility • Siemens Prisma • Siemens Skyra • BIDS • FSL • MRtrix • Neurology 2020