You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I discovered an issue using the CatPhan504 module for our Versa HD linac (not HU calibrated)
The CTP515 ROI values end up being negative, which causes the visibility to be negative, tested by printing the list of:
cp = CatPhan504(cp_folder) contrast_rois = [c.contrast for c in cp.ctp515.rois.values()]
I imagine this is due to scaling/offsets not being applied and pixel values being negative, which just means returning the absolute value for the function;
so that for cases where these offsets are not calibrated on the machine ensures that the Michelson contrast is still bound between [0, 1] for pixel values < 0.
I've also included the dataset used below for reproducibility;
I discovered an issue using the CatPhan504 module for our Versa HD linac (not HU calibrated)
The CTP515 ROI values end up being negative, which causes the visibility to be negative, tested by printing the list of:
cp = CatPhan504(cp_folder)
contrast_rois = [c.contrast for c in cp.ctp515.rois.values()]
I imagine this is due to scaling/offsets not being applied and pixel values being negative, which just means returning the absolute value for the function;
pylinac/pylinac/core/contrast.py
Line 116 in 75e1c2a
I've also included the dataset used below for reproducibility;
dataset.zip
The text was updated successfully, but these errors were encountered: