MTF changes

Hello,

We have recently upgraded from v2.1.0 to v 2.3.2 and have noticed unexpected changes to the MTF analysis.

I can see from the changelogs that the value reported is expected to be half what it used to be. This is true for our Catphan 503 CBCT analysis, running ctp.circle_profile.plot() shows the same lines being selected and the MTF is reduced by a half.

xvi mtf.JPG

For out CT scanner though ctp.circle_profile.plot() shows that a different set of lines are being selected, going from the lines where the trough drops below 50% max to the line set where they are just resolvable.

CT MTF.JPG

The same script is used for the CBCT and the CT so I don’t understand where this change in behaviour is coming from. Can anyone explain what is going on?

Kind regards,

Rosie

Pylinac was previously reporting lines/mm and has been corrected to report line pairs/mm. See: https://github.com/jrkerns/pylinac/pull/271

xvi mtf.JPG

CT MTF.JPG