Input mask for .jpg files

Hey there,
I hope you guys are good, currently I am working on .jpg images, however I’d like to use radiomics.featureextractor.RadiomicsFeatureExtractor() the problem I am having however, is with the .execute is that while I do have the image file path, I don’t have the input mask file, I checked out the jupyter notebooks on github and it seems you only use the .getTestCase method to get both the path and the mask file, so if anyone could help me with that I would be very much grateful :slight_smile:

The getTestCase function only obtains the example cases included in the PyRadiomics repository (downloading them if necessary).

As to your problem, where do you have your mask if you don’t have a filepath? Is it loaded in Slicer? Or do you create the mask dynamically in a Python script? PyRadiomics also supports you to pass the mask a loaded SimpleITK.Image object instead of a file descriptor pointing to the file that can be loaded a SimpleITK.Image object.

Finally, you mention that you work with .jpg input, this is often a color image (i.e. a vector image with 3 values per pixel). PyRadiomics does not support this, input image is required to be grayscale (scalar pixel datatype). The mask is allowed to be a vector image, where you need to select the index to use for your mask.