Hello,

I am having trouble interpreting the Pyradiomics output when applied to PET dicom images, in particular, the values of the ‘diagnostic-original’ features related to the minima, maxima, and range. I am doing the processing without normalization, and I set the default sampling bin width to 25. When I apply this to a large batch of PET dicoms, I get one of two things:

- the minimum value equal to 0, with a max on the order of 18000,
- the minimum value on the order of -32000, and the max on the order of +32000.

This binary dichotomy holds even if I vary my bin width by several orders of magnitude.

Can someone please explain to me what the minima, maxima, range values of my Pyradiomics PET output mean? What are the units of these values and what would cause an image to have one characteristic range or another? I just do not understand what the code is calculating.

Thank you.