CONP Portal | Dataset


Keywords: Phantom MRI
Multicenter Single Subject Human MRI Phantom
Creators: McGill Center for Integrative Neuroscience, Alan C. Evans
Principal Investigator: Alan C. Evans
Licenses: CC BY-ND
Version: 2.0
Formats: MINC JSON
Size: 40.0 GB
No of Files: 3771
No of Subjects: 1
Project Landing Page: https://phantom.loris.ca
Metadata file: DATS.json
Dimensions: resting BOLD, fieldmap BOLD, fieldmap DWI, DWI 65 directions, DWI 25 directions, T1W, T2W
Is About: Homo sapiens, Homo sapiens, adult
Spatial Coverage: North America
Other Dates: Start Date: 2008-04-25 00:00:00 -- End Date: 2019-04-15 00:00:00
Description:
Longitudinal brain scans of a single human phantom acquired on multiple MRI devices across North America over a period of 11 years. In addition to the human brain images, lego phantom scans have been acquired in parallel for quality assessments over time across sites.

Dataset README information

README.md

Multicenter Single Subject Human MRI Phantom Dataset

Overview

This dataset consists of brain scans of a single human phantom acquired on multiple MRI devices
across North America over a period of 11 years. In addition to the human brain images, lego phantom
scans have been acquired in parallel for quality assessments over time across sites.

MRI modalities include: T1-weighted (t1w), T2-weighted (t2w), resting state BOLD (bold),
25-direction DWI (dwi25), 65-direction DWI (dwi65) as well as fieldmaps for the BOLD and
DWI acquisitions (fieldmapBOLD and fieldmapDWI).

The human phantom data were collected once a year during two MRI sessions. Those two sessions were either performed the same day or with a few days apart.

Data organization and file format

Data are organized by subject_ID/Visit_Label.

. 
|__ subject-1
    |__ visit-label-1
       |__ images
           |__ images.json
           |__ phantom_<subject-1>_<visit-label-1>_<scan-type>_<run-number>.mnc
           |__ phantom_<subject-1>_<visit-label-1>_<scan-type>_<run-number>.mnc
           ...
    |__ visit_label_2
    ...
|__ subject_2
|__ subject_3
...
|__ DATS.json
|__ README.md

The ID of the human phantom subject is 963271 and the 9 other subjects present in the dataset are lego phantoms.

Visit labels for the human phantom acquisitions are labelled according to the convention <SiteAbbreviation>_<ScanSessionNumber>_<AcquisitionDate>
Example: MNI_SCAN1_20101125 corresponds to scan session number 1 at the MNI site performed on November 25th, 2010.

The visit labels of the lego phantom acquisitions are labelled according to the convention
<SiteAbbreviation>_<AcquisitionDate>

All images are available in MINC format.

Information about the scanning sites

All MRI scanners were manufacured by SIEMENS.

Abbreviation Name Scanner Model Scanner software Versions
MNI Montreal Neurological Institute TrioTim Syngo MRI B15 & B17
PH1 Philadelphia Children's Hospital MR1 TrioTim Syngo MRI B15 & B17
PH5 Philadelphia Children's Hospital MR5 TrioTim Syngo MRI B17
SEA Seattle Children Hospital TrioTim Syngo MRI B15 & B17
SLC St Louis CIR TrioTim Syngo MRI B17
SLH St Louis Children's Hospital TrioTim Syngo MRI B15 & B17
SLR St Louis Washington University TrioTim Syngo MRI B15 & B17
UMP University of Minnesota Prisma Scanner Prisma fit Syngo MR D13D
UMT University of Minnesota Tim Trio Scanner TrioTim Syngo MR B17
UNH University of North Carolina Hospital TrioTim Syngo MRI B15, B17 & B19
UNR University of North Carolina Research. TrioTim Syngo MR B17 & B19

Acknowledgments

The IBIS Network (https://www.ibis-network.org/) and the NIH (https://www.nih.gov/).

Download Using DataLad

CircleCI status

The following instructions require a basic understanding of UNIX/LINUX command lines. A subset of open datasets on the Portal are also available through a browser-based download button. The instructions below regard dataset download with the use of DataLad. To install DataLad on your system, please refer to the install section of the DataLad Handbook .

Note: For maximum compatibility with conp-dataset, the CONP recommends versions 3.12+ of Python, 10.20241202+ of git-annex, and 1.1.4+ of datalad.

1) Initiate the CONP dataset

Run the following command in the directory where you want the CONP dataset (conp-dataset) to be installed:

2) Install the multicenter-phantom dataset

To install the multicenter-phantom dataset, run the following commands to move into the "projects" subdirectory under the "conp-dataset" directory (created in the previous step) and run datalad install:

3) Download data from the multicenter-phantom dataset

Now that the dataset has been installed, go into the multicenter-phantom dataset directory.

The files visible after installing the dataset but before downloading (in the next step) are symbolic links and need to be downloaded manually using the datalad get command:

If you run datalad get * command, all the files available in the dataset directory will be downloaded.


For more information on how DataLad works, please visit the DataLad Handbook documentation.