-
Notifications
You must be signed in to change notification settings - Fork 1
Home
We discuss software infrastructure for acquisition of MRI data in the c-VEDA project.
Acquisition centres send a single imaging dataset per subject. A dataset shall be uploaded as a ZIP file containing folders, each folder containing a single sequence in the form of anonymized DICOM files. Detailed instructions for operators are collected in a Standard Operating Procedure (SOP) document.
Acquisition centres will run the following MRI sequences during a single neuroimaging session:
| Sequence | Folder name | |
|---|---|---|
| 1 | Scout image | — |
| 2 | Site specific calibration scans, if necessary | — |
| 3 | 3D T1 weighted volume | T1w |
| 4 | Resting state fMRI | rest |
| 5 | B0 mapping, if available | B0_map |
| 6 | DTI | dwi |
| 7 | DTI - reversed | dwi_rev |
| 8 | 2D fast FLAIR | FLAIR |
| 9 | 2D T2 weighted | T2w |
DICOM files from the imaging sessions are packed in a ZIP file. The structure of the ZIP file shall be as follows:
120003456789.zip
├── T1w
├── rest
├── B0_map
├── dwi
├── dwi_rev
├── FLAIR
└── T2w
Acquisition centres use DicomEditor to anonymize neuroimaging data. Data sent to the databank identify subjects by their PSC1 code only. The databank team expect minimal information in specific DICOM tags, depending on the acquisition site:
| ID | Acquisition centre | DICOM tag for PSC1 code |
|---|---|---|
| 12 | CHANDIGARH | (0010,0020) Patient ID |
| 15 | MYSURU | (?) (0040,0280) Comments on the Performed Procedure Step |
| 16 | NIMHANS | (0010,4000) Patient Comments (0032, 4000) Study Comments (single acquisition after RIS) (0010,0020) Patient ID (before RIS) |