review phantom design - Githubissues

Image-X-Institute / MRI_DistortionPhantom

Scripts to generate a FreeCAD model of an MRI distortion phantom

https://acrf-image-x-institute.github.io/MRI_DistortionPhantom/

GNU General Public License v3.0

3 stars 1 forks source link

review phantom design #9

Closed bwheelz36 closed 2 years ago

bwheelz36 commented 2 years ago

Hey Paul

I put a draft of the phantom design on this branch (EvolutionGearPhantom.py)

I made the following changes

adjusted hole size
enlarged DSV to 160 so it is slightly larger than the true DSV
extended phantom length to +- 180 mm - will guide rods be long enough? This is mainly just to make the phantom more flexible if we want to use it in other scanners; I wouldn't propose actually putting markers in the two end slices at this point.
added large load regions for vaseline either side of central slice
saved all marker positions to a text file

Let me know what you think...

paulzyliu commented 2 years ago

I got this building in FreeCAD all ok and it looks good

just a few points

guide rod can be made as long as we want. We didn't test the guide rod fit with the test slice thought right?
enlarging the DSV - given that we are already struggling to get good segmentation for some of the markers at R = 150, do you think this will make the images too hard to segment? (esp for the end slices)
hole depth - I think we want to have all the markers pushed in to the bottom of each hole as that will be the most consistent? do we need to make it shallower based on the test slice?

bwheelz36 commented 2 years ago

I did check the guide rod fits the test slice
DSV. hmmm. yes you're probably right. I'll change it back.
I measured the marker (just with a ruler) and based on that changed the marker depth to 15 mm. Forgot to mention that change. it's probably worth re-doing this with some callipers though,

bwheelz36 commented 2 years ago

another thing to think of is whether we offset the center of the markers from the center of the slice - essentially lower all the markers - to take into account the fact that in most scanners it is difficult to get the markers as low as we can on the MRI-Linac

paulzyliu commented 2 years ago

I did check the guide rod fits the test slice

DSV. hmmm. yes you're probably right. I'll change it back.

I measured the marker (just with a ruler) and based on that changed the marker depth to 15 mm. Forgot to mention that change. it's probably worth re-doing this with some callipers though,

I think we're good to go to order the full phantom then?

paulzyliu commented 2 years ago

another thing to think of is whether we offset the center of the markers from the center of the slice - essentially lower all the markers - to take into account the fact that in most scanners it is difficult to get the markers as low as we can on the MRI-Linac

Yeah this is a good point, we will probably have to do this for the Skyra

bwheelz36 commented 2 years ago

[x] check both Vaseline and markers with calipers
[ ] lower phantom as much as possible

bwheelz36 commented 2 years ago

markers are 15 mm deep confirmed with calipers
added a 10 mm margin for Vaseline.
removed the two end slices I added - we can always get these manufactured later if we need them.
Changed the DSV back to 150
Will make a new issue to enable lowering the phantom

bwheelz36 commented 2 years ago

phantom is lowered by 30 mm compared to previous design.
forced a [0,0] marker on the two end slices

@paulzyliu wanna take a quick look before i contact evolution gear?