Precision of Landmark Positioning on Digitized Models from Patients with Cleft Lip and Palate

Abstract

Objective To quantify the precision of landmark positioning on digitized casts of patients with unilateral cleft lip and palate. Patients Forty plaster models of newborns up to 8 months of age were selected from the archive of the Department of Orthodontics of the University of Heidelberg. Material and Method The plaster-cast models were digitized with a Micromeasure 70 three-dimensional laser scanner (Micromeasure, Bischoffen, Germany). The laser scanner used in this study operates with a precision of 0.15 mm on the x- and y-axes and 0.06 mm on the z-axis. In the intraobserver study, a single observer placed anatomical landmarks in four rounds, with at least 4 weeks between each round. In the interobserver study, four different observers each placed the same landmarks once. For the two different studies, an ideal location for each landmark was calculated by averaging the landmark positions of the four rounds or observers. The distance between each of the four landmark positions and the ideal landmark was measured. Results A 95% confidence interval for the landmark positioning error was calculated. For the intraobserver investigation, this error was 0.34 to 1.30 mm, and for the interobserver investigation it was 0.7 to 2.00 mm. Conclusion Because both investigations displayed comparable error intervals, it was concluded that different observers could perform landmark positioning for the same studies.