Comparison of mandibular landmarks from computed tomography and 3D digitizer data

Area of Science:

• Anthropology
• Medical Imaging
• Biometrics

Background:

• Accurate 3D landmark data is crucial for understanding human mandible development.
• Previous methods for collecting this data were labor-intensive and potentially less precise.

Purpose of the Study:

• To assess the reliability of collecting 3D mandibular landmark coordinates from CT scans using the eTDIPS program.
• To establish a standardized set of landmarks for evaluating craniofacial development and disorders.

Main Methods:

• 3D coordinates of 28 mandibular landmarks were extracted from CT scans of pediatric mandibles (neonate to 13 years) using eTDIPS.
• Direct digitization of the same landmarks on physical mandibles using a Polhemus 3Space digitizer served as the gold standard.
• Linear distances between all landmark pairs were calculated and compared between the two methods.

Main Results:

• eTDIPS showed greater landmark placement variation than direct digitization, with errors concentrated in specific landmark types.
• The mean absolute difference in linear distances between the two methods was 0.377 mm (SD = 1.136 mm).
• eTDIPS-derived coordinates were consistently larger than those from direct digitization.

Conclusions:

• Landmark coordinate data can be reliably collected from digital CT images of the human mandible.
• The defined set of mandibular landmarks is valuable for assessing craniofacial disorders, growth, and other biological processes.
• eTDIPS provides a viable, less invasive method for obtaining precise mandibular measurements.