dke-questions > Tensor metrics differ between FSL_FDT and DKE
Showing 1-3 of 3 posts
Display:
Results per page:
Mar 21, 2018  06:03 AM | minchulusa
Tensor metrics differ between FSL_FDT and DKE
Hello,

I compared the FA images produced by FSL_FDT and DKE
and found that there is somewhat difference.

I give an example of difference image (FDT FA minus DKE FA).

I find it somewhat different at corpus callosum, and which I think should not happen.

I guess is it because of difference in tensor fitting method? 
suppose that FSL do not capture kurtosis term (W)

best wishes
Minchul
Attachment: subtract.jpg
Mar 21, 2018  11:03 AM | Emilie McKinnon - MUSC
RE: Tensor metrics differ between FSL_FDT and DKE
Minchul,

Thank you for your question. Can you tell me what the scale is in your subtract.jpg? Are we talking about a 1% difference or a 10% difference (positive or negative) ? 

Changing the dti_method.dti_flag to 1 in dkeparameters will output an FA map calculated using the DTI method (calculating D without estimating the W term). You can use this map and compare it to the FSL output. 

dti_method.dti_flag = 0; % whether (1) or not (0) to estimate DTI parametric maps based on the DTI (in addition to DKI) signal model (default: 0)

Best, 
Emilie
Jan 27, 2020  10:01 PM | Shereif Haykal
RE: Tensor metrics differ between FSL_FDT and DKE
Hi,

I have a couple of follow-up questions to this old post, if you don't mind:

1. What's the difference between the DTI maps produced by turning on the "dti_method.dti_flag", and those automatically produced with the DKI method using the default settings? For example, what's the difference between the produced maps "fa.nii" and "fa_dti.nii"? 

2. When using the DTI method (with "dti_method.dti_flag" on), are all the DWI shells used to produce the tensor maps or only the lower shell(s)?

I tried to look for answers to my questions in the documentation but couldn't find what I was looking for.

Any help would be appreciated. Thanks.

Best,
Shereif