weird questions

[Ricardo Morais]

Hi Frank, 

1)  

Reconstruction method : 

From the workshop and documentation you've explained the main difference between (GQI) and (QSDR), is that QSDR reconstructs in the template space. For a correlational tractography analysis of a small human group, does QSDR provides any advantage, meaning are the results more reliable that for GQI. I know it depends but would like your insight in it. 

I often get decreased and increased FA findings, which are difficult to understand from a biological point of view. Is the validity of both findings (increase FA and decrease FA) similar ? I'm guessing it is but just to confirm. 

2)

Step C3D: Parameters:

Since I want to keep only “strong” findings and to get ride of noise, I’ve used the following parameters :

FDS ControFDR<0,01

T Threshold: 2,5

Length Threshold: 20 voxels (it’s the parameter that seems to get ride of more “noise” for what I’ve seen, results seem more “cleaner” when I increase the lenght)

Permutation Count: 5000

Pruning Iteration:20

-Are they a good choice Or are they too restricted?  Do you advice to leave the default parameters untouched ?

Thanks,

best wishes for 2023. 

[Ricardo Morais]

Just to add another question:

Should the Motion correction be done before or after top up /eddy correction, or it doesn't matter?
I do get a message that top up /eddy cannot be done after the motion correction, nevertheless it still runs. 

Thanks again,

[Fang-Cheng Yeh]

Just to add another question:

Should the Motion correction be done before or after top up /eddy correction, or it doesn't matter?

After, but

Eddy already includes motion coorrection.

I do get a message that top up /eddy cannot be done after the motion correction,

Yes, it is not recommended.

Best regards

Frank

nevertheless it still runs. 

Thanks again,

A sábado, 7 de janeiro de 2023 à(s) 18:45:04 UTC, Ricardo Morais escreveu:

Hi Frank, 

1)  

Reconstruction method : 

From the workshop and documentation you've explained the main difference between (GQI) and (QSDR), is that QSDR reconstructs in the template space. For a correlational tractography analysis of a small human group, does QSDR provides any advantage, meaning are the results more reliable that for GQI. I know it depends but would like your insight in it. 

I often get decreased and increased FA findings, which are difficult to understand from a biological point of view. Is the validity of both findings (increase FA and decrease FA) similar ? I'm guessing it is but just to confirm. 

2)

Step C3D: Parameters:

Since I want to keep only “strong” findings and to get ride of noise, I’ve used the following parameters :

FDS ControFDR<0,01

T Threshold: 2,5

Length Threshold: 20 voxels (it’s the parameter that seems to get ride of more “noise” for what I’ve seen, results seem more “cleaner” when I increase the lenght)

Permutation Count: 5000

Pruning Iteration:20

-Are they a good choice Or are they too restricted?  Do you advice to leave the default parameters untouched ?

Thanks,

best wishes for 2023. 

--
You received this message because you are subscribed to the Google Groups "DSI Studio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/d88b25b4-9335-41ae-919d-6773b39ffc9fn%40googlegroups.com.