Andy: Update from Wiki

2006-12-18T13:24:18Z

Update from Wiki

New page

'''Scanner:''' 1.5 Tesla GE Echospeed system (General Electric Medical Systems, Milwaukee, WI).

'''Scanning Sequence:''' Maximum gradient amplitudes: 40 mT/M. Six images with 4 high (750 s/mm2), and two with low (5 s/mm2) diffusion weighting. Rectangular FOV (field of view) 220 x 165 mm. 128 x 96 scan matrix (256 x 192 image matrix). 4 mm slice thickness, 1 mm interslice distance. Receiver bandwidth +/-6kHz. TE (echo time) 70 ms; TR (repetition time) 80 ms (effective TR 2500 ms). Scan time 60 seconds/slice.

'''Diffusion Weighted Data:''' The data is stored in nrrd format, the header has a lot of useful information regarding orientation of the scan as well as the gradient orientation in world RAS space. This is not tensor data, it is the raw diffusion data. The data is stored as 7 consecutive volumes (t2w baseline, followed by the gradients) in one binary file:

* HUVA########_dwi.nhdr (ASCII header file)
* HUVA########_dwi.img (256x256xSx7 big endian shorts, S is the number of slices which varies, check the header)

'''Region of Interests:''' ROIs have been drawn for two different structures:

* HUVA########_fornix_l1z150 (256x256xS big endian shorts)
* HUVA########_cingulum_bundle (256x256xS big endian shorts)

The description of the labels and corresponding structures is in [[Media:Diffusion-ROI-description.xls| [Diffusion-ROI-description.xls]]].

'''Important Note: All but 4 cases were scanned in the Coronal PA order. These 4 cases were scanned in Coronal AP, they are:'''

* HUVA02120485
* HUVA30283926
* HUVA27418186
* HUVA62180131

'''These four cases will not correctly load with the same gradients as the other studies.'''

''''''

Here is a ppt presentation about nrrd and its integration with ITK/slicer [[Media:Nrrd-namic2.ppt|Nrrd-namic2.ppt]].

Here is a [http://teem.sourceforge.net/nrrd/namic.html page with info and links about NRRD format usage]

Brockton DTI Scan Parameters - Revision history

Andy: Update from Wiki