slicer radiomics and script different feature values

[Namaste Tenzin]

Hi Joost,

I have been getting different values from slicer 3d and I used resampling pixel 1,1,1 when I loaded the parameter using the scrip with resampling pixel 1,1,1 and siktBspline, I get different feature value from slicer 3d and running the script on pycharm.

I used Mac to run slicer 3d and it runs on python 2.7 where as pycharm on python 3,6 could that be the reason?

so the feature value from slicer 3d from Mac on python 2.7 is wrong?

best

Tenzin 

[Joost van Griethuysen]

Hello Tenzin,

Which versions of PyRadiomics are you using? What other kind of customization? Do you have a log file of the extraction you can share?

Whether python 2.7 or python 3.6 is used should not change the output.

Regards,

Joost

Op donderdag 14 november 2019 07:04:01 UTC+1 schreef Namaste Tenzin:

[j.v.gri...@nki.nl]

Hi Tenzin,

 

You can export the table from slicer to a csv file, and you can store PyRadiomics results as a csv file too. Then, you can read (and compare) them in python using pandas. That’ll be quicker, less tedious and less error-prone.

 

Regards,

 

 

Joost van Griethuysen

 

From: Namaste Tenzin [mailto:tkun...@gmail.com]
Sent: donderdag 14 november 2019 21:12
To: Joost van Griethuysen
Subject: Re: slicer radiomics and script different feature values

 

Hi Joost,

 

i am using 2.2 version on both 3d slicer on mac and the script. perhaps i will check again if the values are same or not...its tedious to check each values one by one since the order of feature output is different in slicer 3d and on script. 

 

best

tenzin 

 

--
You received this message because you are subscribed to the Google Groups "pyradiomics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pyradiomics...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pyradiomics/d80c220a-bed9-4c5e-a3c4-2fe956c3592e%40googlegroups.com.

[Namaste Tenzin]

Thanks for the message Joost,

Although I am not sure why but generally with slicer 3d and running script on pycharm its the same values but for some patients I get different radiomic values. I will send you the test log here but for some patients the value differs and I am not sure why, I used the exact same setting. 

please let me know what went wrong. 

best

Tenzin