Evaluating the compatibility of the 2D and 3D facial soft tissue depth measurement methods

Forensic facial approximation is a technique that involves constructing the facial muscles and applying a suitable facial soft tissue depth (FSTD) dataset. To date, several FSTD studies have been conducted for varying population groups by using different FSTD measurement methods (i.e., needle puncture, two-dimensional (2D) and three-dimensional (3D) FSTD measurement methods). This study aims to compare the 2D and 3D FSTD measurement methods. The facial depths of subjects were measured using 100 head CT scans of 50 male and 50 female subjects aged between 18 and 99. For the 3D method, the 3D head and skull models of individuals were created by using the Histogram method to segment the CT images in Amira 6.1. Subsequently, the 3D depth measurements were obtained from the 3D head and skull models at 15 (4 mid-sagittal and 11 bilateral) cranial landmarks. For the 2D method, the FSTD measurements were directly taken from the CT slices at the same cranial landmarks as the 3D method. The intra- and inter-observer errors of the measurements were assessed using a repeated measures MANOVA. Similarly, the discrepancies between the FSTD datasets obtained using the 2D and 3D methods were evaluated using a repeated measures MANOVA. The depth differences between the two FSTD datasets were statistically significant, yet they were negligible from a practical point of view.