Ability to Generate a Mean Projection Image Using only Selected Slices Based on Fiducial Markups

Hello,

I need to markup many landmarks on head MRIs with the intent for them to be used like a lateral ceph radiograph is currently used for orthodontic treatment planning.

I can visualize/place the markers on the MRI scans in the individuals panes, but when I go to condense the images into a single 2D-image, I lose the anatomical detail of landmarks I need. I can scroll through the stack and see them, but the goal is to have a single 2D image to analyze. I’ve played around with the code to get the mean projection image (slab numbers/slice thickness) based on code Andras provided in another post, and I’ve adjusted contrasts, etc, but the image is still too ill-defined.

Is there a way I can combine the image slices specifically around my fiducials into one mean projection image and ignore the other slices? I think that may sharpen the landmarks I need to see. I’m also open to other ideas/ways to improve the image quality.

For reference, code from Andras for mean projection image generation:
sliceLogic = appLogic.GetSliceLogic(sliceNode)
sliceLayerLogic = sliceLogic.GetBackgroundLayer()
reslice = sliceLayerLogic.GetReslice()
reslice.SetSlabModeToMax()
reslice.SetSlabNumberOfSlices(500)
reslice.SetSlabSliceSpacingFraction(0.1)
sliceNode.Modified()

Picture from internet showing what a lateral ceph for ortho treatment planning looks like and what I am trying to obtain similarity to (with the understanding the colors/bone will be inverted):


(Taken from: Durão, Ana & Morosolli, Aline & Pittayapat, Pisha & Bolstad, Napat & Pinhão Ferreira, Afonso & Jacobs, Reinhilde. (2015). Cephalometric landmark variability among orthodontists and dentomaxillofacial radiologists: A comparative study. Imaging Science in Dentistry. 45. 213. 10.5624/isd.2015.45.4.213. )

My images. Note: I have many more markers to place - these are just to insure I can get this to work.

Close up of the best quality mean projection image I was able to get using the above code
image

Thanks in advance for your assistance. This community is great!

1 Like