search

jrkerns / pylinac

An image analysis library for medical physics
https://pylinac.readthedocs.io/en/latest/
MIT License
153 stars 98 forks source link

Number of failed ROIs in LeedsTOR analysis does not match between analyzed image and plot #194

Closed mchamberland closed 5 years ago

mchamberland commented 5 years ago

Describe the bug The number of failing low-frequency contrast ROIs shown on the LeedsTOR analyzed image does not match the number of failing ROIs on the plot of the low-frequency contrast.

To Reproduce Run the LeedsTOR analysis on my uploaded image.

Expected behavior The number of failing ROIs on the analyzed image should match the number of failing ROIs on the plot.

Screenshots leeds

Additional context I've uploaded my image to Dropbox.

mchamberland commented 5 years ago

Hmm... actually, maybe I just don't understand what's being plotted!

I'm reading the source and documentation more carefully right now to try to make sense of those results.

mchamberland commented 5 years ago

Okay, my confusion stems from the fact that the analyzed image shows the low contrast ROIs that fail the contrast threshold, whereas the plot is showing the contrast constant values.

Given that the two other phantoms in the planar imaging module use the contrast constant results on the analyzed image, would it make sense to also use the contrast constant for the Leeds phantom?

See here for Leeds.

and compare to Las Vegas here.

In other words, L868 for the Leeds phantom should be:

roi.plot2axes(img_ax, edgecolor=roi.plot_color_constant)

jrkerns commented 5 years ago

This was a test and not a mistake in any way. 😑 You passed.