You must be logged in to post Login Register


Lost Your Password?
Search Forums:


 






Wildcard Usage:
*    matches any number of characters
%    matches exactly one character

Forum Diffusion Toolkit Usage how to make a "mgh_dti_025" for DTI data with 25 directions?

how to make a "mgh_dti_025" for DTI data with 25 directions?
Page:   1  
UserPost

12:00 pm
July 18, 2010


meikei

Member

posts 5
Print this Post
1

Hi all,

The number of
directions of my DTI data is 25, but I can't find a suitable gradient
table in the Diffusion ToolKit to fit it. I wonder where can I download / how can I create a "mgh_dti_025"?

Thanks very much for any of your kind help!

 

Meikei

10:04 am
July 19, 2010


meikei

Member

posts 5
Print this Post
2

Hi all,

I found that DTK allows me to use a gradient table with fewer number of directions (e.g. mgh_dti_006 or 018, as my data has 25 directions) to import my DTI data, but not a gradient table with larger number of directions (like mgh_dti_030). I tried to run the DTK with 006 and 018. However, the track file "dti" automatically generated after the run looks quite weird (see below)… as I think there should be tracks all over the brain normally? Welcome any comments, and thanks for any help.

 

Meikei

 

12:30 pm
July 19, 2010


Ruopeng

Admin

posts 353
Print this Post
3

Hi Meikei,

You can't just pick a random gradient table to reconstruct your data. You need to make sure to use the directions that were used to collect your data. As I posted in another thread, if your sequence is MGH's, here is the 25 directions:

  1.0000000000  0.0000000000  0.0000000000
  0.7399404420 -0.6050663212  0.2939096613
  0.7398746493  0.3801861874 -0.5550170863
 -0.4424123013 -0.2143947312 -0.8708078174
  0.3115437133 -0.4142049916  0.8552045016
 -0.3727450012 -0.2352033215  0.8976305261
  0.7312921455  0.6466417238  0.2169476408
 -0.9199013765 -0.2560014116 -0.2970601532
 -0.0312165063 -0.8818785508  0.4704420807
  0.1107650856 -0.6487074769 -0.7529340643
 -0.0312555219  0.5957471270 -0.8025636753
  0.1069858839  0.8775254702  0.4674431191
 -0.8873541059 -0.0572050375  0.4575262553
  0.3114214593  0.9071107056 -0.2831374971
  0.7312365937  0.1191121812  0.6716437539
 -0.6588285365  0.4019529617 -0.6359078361
  0.1068419113  0.3324156892  0.9370616924
 -0.8874090021  0.4608510654  0.0110253512
 -0.4423300948 -0.8295927562 -0.3407637684
  0.7544231925 -0.4283907318 -0.4973198441
 -0.6588388622 -0.6883468868  0.3034961564
 -0.5187105448  0.5580806573  0.6476768875
  0.2903105700 -0.9469309724 -0.1379909645
 -0.3728091932  0.9225341602 -0.0997197497
  0.2901669640  0.0036040713 -0.9569692490

However, if your sequence is not the MGH one. The directions are very likely different. You need to check with the person who knows more information about the sequence you used.

Ruopeng

11:24 pm
July 19, 2010


meikei

Member

posts 5
Print this Post
4

Hi Roupeng,

 

Thanks for your reply. Though I'm not sure about the sequence that was used to collect the data, I guess there's still something wrong as the "dti" just created with the 25 directions you gave still looks like the picture I posted previously.

I will email to ask about the sequence. But does the brand of scanner matter? I used a GE scanner.

Thanks very much for your kind help!

 

Meikei

 

 

2:27 pm
July 20, 2010


Ruopeng

Admin

posts 353
Print this Post
5

If its GE, it most likely is not the same as MGH one (for Siemens scanners).

The reason you only see a slice of tracks is because TrackVis by default will apply a slice filter when the track file is big. You should take a look at the onlne demos here http://trackvis.org/demo for how trackvis works.

Best,

Ruopeng
Page:   1