A comparative reliability analysis of computer-generated bitemark overlays

• Published in: Journal of forensic sciences Vol. 50; no. 2; p. 400
• Main Authors: McNamee, Anne H, Sweet, David, Pretty, Iain
• Format: Journal Article
• Language: English
• Published: United States 01.03.2005
• Subjects: Bites, Human; Computer Simulation; Dentition; Forensic Medicine - methods; Humans; Observer Variation; Reproducibility of Results; Software
• ISSN: 0022-1198
• DOI: 10.1520/JFS2004206

This study compared the reliability of two methods used to produce computer-generated bitemark overlays with Adobe Photoshop (Adobe Systems Inc., San Jose, CA). Scanned images of twelve dental casts were sent to 30 examiners with different experience levels. Examiners were instructed to produce an overlay for each cast image based on the instructions provided for the two techniques. Measurements of the area and the x-y coordinate position of the biting edges of the anterior teeth were obtained using Scion Image software program (Scion Corporation, Frederick, MD) for each overlay. The inter- and intra-reliability assessment of the measurements was performed using an analysis of variance and calculation of reliability coefficients. The assessment of the area measurements showed significant variances seen in the examiner variable for both techniques resulting in low reliability coefficients. Conversely, the results for the positional measurements showed no significant differences in the variances between examiners with exceptionally high reliability coefficients. It was concluded that both techniques were reliable methods to produce bitemark overlays in assessing tooth position.