Diffusion MRI: Difference between revisions
| Line 28: | Line 28: | ||
| 2.0x2.0x2.0 | | 2.0x2.0x2.0 | ||
| ?? | | ?? | ||
| {{hidden | ta1 = left | headercss = font-size: 100%; width: | | {{hidden | ta1 = left | headercss = font-size: 100%; width: 24ex; | contentcss = font-size: 90%; width: 24ex; | ||
| header = Not | | header = Not processed | ||
| content = Single dataset needs to be processed ASAP for social reasons | | content = Single dataset needs to be processed ASAP for social reasons | ||
}} | }} | ||
| Line 42: | Line 42: | ||
| ?? | | ?? | ||
| ?? | | ?? | ||
| {{hidden | ta1 = left | headercss = font-size: 100%; width: | | {{hidden | ta1 = left | headercss = font-size: 100%; width: 24ex; | contentcss = font-size: 90%; width: 24ex; | ||
| header = Not downloaded | | header = Not downloaded | ||
| content = Longitudinal data and old proof-of-concept data need to be loaded into <code>$G/data/diffusion</code> | | content = Longitudinal data and old proof-of-concept data need to be loaded into <code>$G/data/diffusion</code> | ||
| Line 55: | Line 55: | ||
| ?? | | ?? | ||
| ?? | | ?? | ||
| {{hidden | ta1 = left | headercss = font-size: 100%; width: | | {{hidden | ta1 = left | headercss = font-size: 100%; width: 24ex; | contentcss = font-size: 90%; width: 24ex; | ||
| header = Not collected | | header = Not collected | ||
| content = Still waiting for the 7T protocol to finalize | | content = Still waiting for the 7T protocol to finalize | ||
Revision as of 15:55, 19 August 2008
Collaborative Research Projects
| Project | Location | Collaborators | Dataset Date Received / Path | Scanner | Format | Resolution (voxels) | Voxel Size (mm) | Orientation | Status |
|---|---|---|---|---|---|---|---|---|---|
| Cognitive Aging | St. Louis | Rob Paul (UMSL), Tom Conturo (WUSTL) | 2008-07-20
|
?? | Interfile | 83x110x64 x 36 DWIs | 2.0x2.0x2.0 | ?? | Not processed
|
| Fibers at Risk | Colorado | Jack Simon | unknown / not transferred | 1.5T ?? | ?? | ?? | ?? | ?? | Not downloaded
|
| Portland, OR | Jack Simon (Portland VA), Bill Rooney (phys. chemist at OHSU AIRC) | not transferred | 7T GE | ?? | ?? | ?? | ?? | Not collected
|
Software
File Formats
DICOM
DICOM Mosaic
Analyze
NIfTI
Files ending in .nii are in the NIfTI medical image format; it is becoming a standard export format for many different software packages, including FSL. .nii files combine the metadata header and image data into a single file, but NIfTI-1 is a modified form of Analyze 7.5, so it sometimes comes in .hdr/.img pairs. You can test for the format by using the is_nifti_file() function in the I/O wrapper library.
The format was created by the Neuroimaging Informatics Technology Initiative at NIH. It is a superset of Analyze 7.5, re-purposing empty fields in the Analyze format to support more features. Fortunately for us, the NIfTI reference library is available in the public domain, and $G/src/nifti is set up to download the latest version (with make download), build it, and install it in $G/lib/nifti and $G/include/nifti. You might also try snooping around $G/doc/nifti to see what they have to say in their documentation. $G/src/nifti2mriimage/nifti2mriimage.cpp demonstrates reading the header and extracting information from it, as well as reading the image data.
To use the NIfTI I/O wrapper library, simply #include <nifti1_io.h> in your code and add the following to the library list in your makefile: niftiio znz m z.
From the command line or shell scripts, the nifti_tool utility is pretty powerful and can read the header in a structured way.
The online documentation for NIfTI is also pretty sweet:
Interfile
MRIimage
MRIimage is our in-house format for storing MRI data. It is manipulated by the gg/mri_g++-* library and its header is <gg/mri/image.H>.