Date of Completion


Embargo Period



Aditya Tadinada, Sumit Yadav, Ravindra Nanda, Flavio Uribe

Field of Study

Dental Science


Master of Dental Science

Open Access

Campus Access


Traditional two-dimensional (2-D) cephalometric methods suffer from several inherent errors. With the introduction of Cone-beam Computed Tomography (CBCT), an accurate three-dimensional (3-D) representation of the craniofacial form could be generated. Objectives: The overall objective of this pilot study was to overcome limitations of traditional 2-D cephalometry by developing and testing a protocol for generating an average 3-D human skull using CBCT images. Specific goals were 1) To identify landmarks and regions on the skull appropriate for geometric morphometric analysis especially in relation to orthognathic surgery patients, 2) Develop protocol for applying geometric morphometric techniques to generate an accurate 3-D average skull model suitable for surgical orthodontic applications, 3) To compare the gender differences between the male and female average skull models, and 4) To explore the usefulness of the normalized skull models for diagnosis and evaluation of treatment outcomes. Methods: Thirty CBCT scans obtained from normodivergent adults with Class I skeletal pattern were used. Geometric morphometric analysis of these scans was performed with the Checkpoint software (Stratovan Corporation, Davis, CA). Eighteen landmarks and 8 patches with semi-landmarks encompassing a total of 666 points were identified on each 3-D surface model generated from the CBCT scan. Once homologs points were collected statistical shape analysis was performed using absolute averaging and Procrustes superimposition technique. Once population averages were computed, normalization was performed on two specimens with Class II and III skeletal malocclusions. Results: Simple linear measurements computed from 3D coordinate data revealed that males had a significantly larger anterior cranial base length, transverse dimensions and ramus height. Two 3-D skull models were generated for each gender based on the aforesaid absolute and Procrustes averaging techniques. Normalization of 3-D skulls with Class II and III deformities enabled the identification of most affected regions in the skull that needs to be corrected by orthognathic surgery. Conclusion: The constructed average 3-D skull models were a viable alternative to the traditional 2-D cephalometry. These models can be used for diagnosis, treatment planning and evaluating treatment outcomes.

Major Advisor

Aditya Tadinada