Winston Lutz BB distance to EPID

Google Groups
1

Hello,

I guess I’m a trouble maker.
So I’m doing Winston Lutz on a Halcyon for SRT…
(working with what we have at the moment… please no comments from the peanut gallery)
The field edge defined center seemed to be off a bit from EPID - so my chief asked to get values of BB to EPID based center instead of BB to CAX as defined by field edge.

Thought there would be a quick flag to choose from but there wasn’t.
Is anyone else interested in this? My chief’s justification was he wanted to just test IGRT and placement of BB in MV fields in his Winston Lutz and not also add the complexity of the field edge from the MLCs.

Example 1.13 mm reduced to 0.69 mm.

My quick cheater’s way to calculate this
was to replace the definition of CAXtoBB with EPIDtoBB. This probably will screw with some calculations but I will also probably reduce all the other info outputs to only show max/median EPID to BB distance and the graphs (with EPID to BB shown):

def cax2bb_vector(self) → Vector:
“”“The vector in mm from the CAX to the BB. #JSH “””
dist = (self.bb - self.epid) / self.dpmm
return Vector(dist.x, dist.y, dist.z)def cax2bb_distance(self) → float:
“”“The scalar distance in mm from the CAX to the BB. #JSH “””
dist = self.epid.distance_to(self.bb)
return dist / self.dpmmversus Original:

def cax2bb_vector(self) → Vector:
“”“The vector in mm from the CAX to the BB.”“”
dist = (self.bb - self.field_cax) / self.dpmm
return Vector(dist.x, dist.y, dist.z)def cax2bb_distance(self) → float:
“”“The scalar distance in mm from the CAX to the BB.”“”
dist = self.field_cax.distance_to(self.bb)
return dist / self.dpmm

2

Hi there,

If you are interested in BB-EPID distance, you can simply subtract the cax2bb and cax2epid vectors. That should do the trick, I think.

On another note … although it may be interesting to observe the bb2epid distance, I would strongly discourage you to do that if your intention is to check the IGRT system. This distance has a proper meaning only if you are testing the “positional” calibration of the epid, ie how the center of the epid is aligned with the CAX. Which, I think, in the newer Varian system is what counts.
The imaging software, on the other hand, uses additional corrections that come from the calibration itself. So be careful. Your bb2epid distance will be the distance between the BB and the epid center, not the “isocenter” as seen by the software.

I suggest that you turn the story around: first set the BB into the isocenter with imaging (CBCT). Do it in such a way that the positional error, as shown by the IGRT software, is 0. Then acquire MV images and observe cax2bb. This will give you the exact error vector of your treatment in ideal circumstances.

Best regards,
Denis