Questions regarding tract-specific analysis from the commandline

[Donnie Cameron]

Dear Frank,

I hope you are well. You may remember assisting me with some muscle DTI questions in the past. Thanks again for your help!

I have a few questions about tract-specific analysis, if that's OK. I've got a large number of datasets, each with many regions, for which I'd like to extract some tract-specific measures using the DSI Studio commandline; however, I see that the "-action=ana, --export=stat" approach only exports a limited set of measures (tract lengths, mean FA, mean ADC, etc.). My first question is, is there a way I can extract custom tract-specific measures? In particular, I would like to measure the median FA and ADC instead of the mean, and I'd like to export values for the second and third eigenvalues (radial dif1 and radial dif2) as well. Is this possible?

Secondly, is there a way to combine .trk files in the DSI Studio commandline and get statistics from the combined tracts? I have multiple .trk files representing individual muscles, and I'd like to combine these to get stats for whole groups of muscles (quadriceps, hamstrings, etc.). I could, in theory, perform fibre-tracking again over the group of ROIs, but this will take a good deal of extra time and will produce some very large files. Do you have any workarounds for this?

I guess both of these questions could be cleared up if I could access the data in the .trk files directly, and perform my analysis on these data. Is this possible and, if so, are there any instructions for how to do this? I'd be grateful for any insights you might be able to share.

With gratitude,

Donnie         

[Fang-Cheng Yeh]

I have a few questions about tract-specific analysis, if that's OK. I've got a large number of datasets, each with many regions, for which I'd like to extract some tract-specific measures using the DSI Studio commandline; however, I see that the "-action=ana, --export=stat" approach only exports a limited set of measures (tract lengths, mean FA, mean ADC, etc.). My first question is, is there a way I can extract custom tract-specific measures? In particular, I would like to measure the median FA and ADC instead of the mean, and I'd like to export values for the second and third eigenvalues (radial dif1 and radial dif2) as well. Is this possible?

--action=ana --export=fa,adc  will take along tack FA and ADC value into text files, and then you can load values in these text files to MATLAB to calculate the median.

--action=rec --output_tensor=1 will output rd1 and rd2 in addition to FA and ADC...etc.

 

Secondly, is there a way to combine .trk files in the DSI Studio commandline and get statistics from the combined tracts? I have multiple .trk files representing individual muscles, and I'd like to combine these to get stats for whole groups of muscles (quadriceps, hamstrings, etc.). I could, in theory, perform fibre-tracking again over the group of ROIs, but this will take a good deal of extra time and will produce some very large files. Do you have any workarounds for this?

I will revise the code to make this possible. something like --action=ana --source=my.fib.gz --tract=1.trk.gz,2.trk.gz,3.trk.gz

Once available, I will let you know.

 

Best,

Frank

--
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/4bca66ca-9da5-4c70-a531-0c212f8e5fbbo%40googlegroups.com.

[Fang-Cheng Yeh]

Are you using Windows or the Mac version?

Frank

[Donnie Cameron]

Hi Frank,

Thanks very much for the incredibly fast and helpful reply. Much appreciated!

In answer to your question - I'm using the Windows version of DSI Studio.

Many thanks,

Donnie 

To unsubscribe from this group and stop receiving emails from it, send an email to dsi-s...@googlegroups.com.

[Fang-Cheng Yeh]

I committed an update to the Windows version (dated today afternoon around 5 pm). Please check if it works for you. 

You can now assign multiple files at --tract=

DSI Studio will process each of them separately and output separate results.

If you want to combine the result, a quick way is to do a volume-weighted average:

combined FA = (V1xFA1 + V2xFA2)/(V1+V2)

Best,

Frank

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/1aae2b73-4239-4faf-a0a3-869ab7cc5388o%40googlegroups.com.

[Donnie Cameron]

Hi Frank,

Thanks for implementing this so quickly. It works great!

Ideally I would prefer to combine the tracts from the different files - so I could view the distribution of diffusion parameters with a histogram, for example. I guess I would still have to do this the 'hard' way, but your modifications do make it easier. Much appreciated!

Kind regards,
Donnie

On Tuesday, 6 October 2020 23:37:19 UTC+2, Frank Yeh wrote:

I committed an update to the Windows version (dated today afternoon around 5 pm). Please check if it works for you. 

You can now assign multiple files at --tract=

DSI Studio will process each of them separately and output separate results.

If you want to combine the result, a quick way is to do a volume-weighted average:

combined FA = (V1xFA1 + V2xFA2)/(V1+V2)

Best,

Frank

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/1aae2b73-4239-4faf-a0a3-869ab7cc5388o%40googlegroups.com.