Three-dimensional facial anatomy evaluation: Reliability of laser scanner consecutive scans procedure in comparison with stereophotogrammetry

Abstract

Laser scanner devices are acquiring a growing importance in facial anatomy. Most studies have analysed facial scans obtained through two simultaneous captures, whereas the same result can be obtained by consecutive three-dimensional (3D) scans. However, this latter procedure has not yet been validated. Fourteen volunteers underwent face image capture through stereophotogrammetry (VECTRA M3) and three consecutive facial scans through a laser scanner (Konica Minolta Vi910). The concordance between 14 linear distances, 12 angles, facial surface area and volume measurements was verified by the Bland-Altman test and calculation of absolute and relative technical errors of measurement (TEM/rTEM). The two facial images obtained by the different devices were then registered to calculate point-to-point distance. Most of linear distances and angles showed a high agreement, with "very good" or "good" rTEMs, ranging between 1.1% and 6.4%. Surface area measurements agreed well between the devices (rTEM: 6.3%), while volumes were poorly comparable (rTEM: 25.8%); the root mean square point-to-point distance was 0.80 mm (SD: 0.41). This study first tested the concordance of measurements on facial images obtained by stereophotogrammetry and consecutive laser scans. Results highlight the reliability of linear distances, angles and surface areas measurements, but discourage volume assessment and registration of surfaces acquired through different devices.

Keywords: Face; Laser scanner; Photogrammetry; Root mean square (RMS).