Winston-Lutz report individual collimator positions

[Alistair Templeton]

I was hoping to use some winston-lutz images to evaluate the slop in jaws/MLC by comparing the left and right collimators in gantry 90/270.  Are the individual distances from ball-to-edge ever calculated in a way that's accessible?

I reckon just looking at normally-reported x/z values between 90 and 270 should give the average slop between the left and right collimator, but if I can pull out the both that'd be swell!

[tbe...@gmail.com]

Hi,

here is a link to the Algorithm: https://pylinac.readthedocs.io/en/latest/winston_lutz.html#implementation

As far as I understand it, the CAX is not defined as the center of the field edges but as the center of mass from a white patch in a black and white image. So the distances to individual collimators are not calculated.

[Alistair Templeton]

Much smarter to rtfm rather than dig through the code. That makes sense, thanks so much!