NCJ Number
225903
Journal
Journal of Forensic Sciences Volume: 54 Issue: 1 Dated: January 2009 Pages: 177-184
Date Published
January 2009
Length
8 pages
Annotation
This pilot study evaluated a computer-based method for comparing digital dental images, using a registration algorithm to correct for variations in projection geometry between images prior to a subtraction analysis.
Abstract
The results showed that at a threshold of 0.855, there were 2 false negative and 2 false positive identifications out of 957 analyses. Analysis of variance determined that there was a significant difference between images from the same individual and those from different individuals. Findings further showed that the threshold level of concordance varied with the anatomical region of the mouth examined. This method may provide the most objective and reliable method for postmortem dental identification using intraoral images. The study involved the digitization of 2 sets of full mouth intraoral radiographic films from 25 patient records, using an Epson Expression 16 Professional flatbed scanner with transparency adapter. All films were digitized at 400 dpi and 8 bit (256) gray levels. Patient information was redacted from the selected radiographs for patient confidentiality and random patient numbers assigned to each image. All image registration and analysis was performed on an Intel Pentium II personal computer using UTHSCSA ImageTool version 3.0 and the UT-ID plug-in module. Once the four anatomical points were chosen in the postmortem and antemortem image, the UT-ID algorithm for scaling, translation, rotation, and perspective distortion of identical structure was applied to the antemortem image. Means of the UT-ID values for similar versus nonsimilar images were compared using a Student’s t-test. Sensitivity, specificity, false acceptance rates, and false rejection rates were calculated for each region examined and for all regions combined. 9 tables, 4 figures, and 36 references