Identification of Stable Craniofacial Landmarks for Superimposition

Objectives: With the advent of rapid scanning, low-radiation systems for the head, the ability to now define craniofacial growth and treatment in three dimensions with volumetric data is developing. The field needs to define how three-dimensional data can be analyzed based on morphometric parameters. This study located stable, localized craniofacial landmarks in which superimposition of a pair of CBCT images could be performed for the purpose of evaluating growth and orthodontic/orthognathic treatment.

Methods: Data were taken from CBCT scans using the Hitachi MercuRay system (Hitachi, Tokyo, Japan) and software to develop both radiographic and three-dimensional volumetric data (CB Works 2.0, Seoul, Korea). Patients who had multiple scans taken more than a year apart were selected for this study. Using an advanced multi-processing and volumetric image program (Amira version 4.1.1, Mercury Computer Systems, Boston, MA), craniofacial landmarks were identified in each pair of scans. Five landmarks were identified (pogonion, gnathion, menton, and the left and right mental foramina). A least squares algorithm was used to determine the translational and rotational parameters required for superimposition. The distance between each vertex of the first surface and the closest point on the second surface was computed with Amira's SurfaceDistance module.

Results: The accuracy of superimposing on the landmarks was assessed visually with a color map. Most pairs of mandibles were aligned with less than 1mm distance between the two surfaces. Other structures in the head were less accurately aligned.

Conclusions: Superimposition on the aforementioned landmarks provides a quick and reasonably accurate visualization of where growth/changes occur in mandibles. Other landmarks are required to better assess growth across the craniofacial region.