Accuracy and repeatability of anthropometric facial measurements using cone beam computed tomography

The purpose of this study was to determine the accuracy and repeatability of linear anthropometric measurements on the soft tissue surface model generated from cone beam computed tomography scans.

The study sample consisted of seven cadaver heads. The accuracy and repeatability were assessed by means of a series of 21 standardized, linear facial measurements derived from 11 landmarks taken both directly on the face with a set of digital calipers and indirectly from a three-dimensional soft tissue surface model generated from a cone beam computed tomography scan of the heads using SimPlant® Ortho Pro software. The landmarks and measurements were chosen to cover various regions of the face with an emphasis on the oral-nasal region. The cone beam computed tomography measurements were compared with the physical measurements. Statistical analysis for the repeatability was done by means of the intraclass coefficient. Accuracy was determined by means of the absolute error and absolute percentage error.

The cone beam computed tomography measurements were very accurate when compared with the physical measurements (0.962 to 0.999). Except for one measurement, between point tragion (t) and nasion (n) (mean, 1.52 mm), all the measurements had a mean absolute error of less than 1.5 mm.

The three-dimensional surface models derived from cone beam computed tomography images are sufficiently precise and accurate for the anthropometric measurements.