Since no one shared their thoughts, I will give you my opinion. The issue is not the lack of bony landmarks, it is the fact that you are trying to measure the shape of a collapsible tube. In addition to the problems you identified (i.e change due to head and tongue position), I suspect the shape will also change where they are in breathing cycle. And then the fact that lateral cephalograms are 2D projections of a 3D complex shape. Normal approaches of doing repeated measurements to reduce error will not help since all of those factors are settled by the time cephalogram is taken. You need multiple exposures, which it is not easy to ethically defend for a research project like this due to ionizing radiation.
Couple things I would try before starting. See if you have any patients with multiple images. Perhaps technician wanted retake the image due to patient movement or some other concerns. That way you might at least get a sense of what kind of variability you are dealing with and whether it would be meaningful to spend time on the project. Another possible alternative is to use CBCTs, and extract reconstructed lateral cephalograms from them. Then you can keep the projection plane the same but tilt the head a few degrees in 3D to simulate different poses. On Saturday, September 5, 2026 at 4:16:01 PM UTC-7 Sushant Pandey wrote: > Hi all, > > I'm an early-career orthodontic researcher, new to this group. Can GMM be > validly applied to a structure like the pharyngeal airway space as seen on > a lateral cephalogram? The space has no bony landmarks, defined entirely by > soft-tissue boundaries, and potentially affected by head position and > tongue posture at the time of imaging? > > Would appreciate any references or experience with this. > > Thanks, > > Sushant Pandey > -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/morphmet2/1a57fa28-aec0-4eb7-8f7a-1c5986069a9an%40googlegroups.com.
