Computer‐aided Dental Identification: An Objective Method for Assessment of Radiographic Image Similarity*

Abstract

A pilot study evaluated a computer-based method for comparing digital dental images, utilizing a registration algorithm to correct for variations in projection geometry between images prior to a subtraction analysis. A numerical assessment of similarity was generated for pairs of images. Using well-controlled laboratory settings, the method was evaluated as to its ability to identify the correct specimen with positive results. A subsequent clinical study examined longitudinal radiographic examinations of selected anatomical areas on 47 patients, analyzing the computer-based method in making the correct identification based upon a threshold level of similarity. The results showed that at a threshold of 0.855, there were two false negative and two false positive identifications out of 957 analyses. Based on these initial findings, 25 dental records having two sets of full mouth series of radiographs were selected. The radiographs were digitized and grouped into six anatomical regions. The more recent set of films served as postmortem images. Each postmortem image was analyzed against all other images within the region. Images were registered to correct for differences in projection geometry prior to analysis. An area of interest was selected to assess image similarity. Analysis of variance was used to determine that there was a significant difference between images from the same individual and those from different individuals. Results showed that the threshold level of concordance will vary with the anatomical region of the mouth examined. This method may provide the most objective and reliable method for postmortem dental identification using intra-oral images.

Root Morphology and Anatomical Patterns in Forensic Dental Identification: A Comparison of Computer‐Aided Identification with Traditional Forensic Dental Identification*

“…Mean UT‐ID Index scores for the “True Positive” pairs was 0.763 (SD 0.135) and for the “True Exclusion” pairs was 0.427 (SD 0.225). Using the UT‐ID Index threshold levels for positive identification as suggested by Flint (17) (see Table 7), sensitivity was 53.8% and specificity was 100%. However, considering the proposed threshold for positive identification for mandibular premolars is 0.650 and that for mandibular molars is 0.800 and given the close anatomical proximity of these regions such that a posterior mandibular periapical radiograph could contain both premolars and molars, when the proposed threshold for both mandibular areas is set at 0.700, sensitivity is 76.9% (three fewer false negatives) while specificity remains unchanged at 100%.…”

“…UT‐ID generates a normalized coefficient of similarity (UT‐ID Index), which is based on the CCC. Flint (17) determined that there was a significant difference in UT‐ID Index values between images taken at different times from the same individual and those from different individuals, and that the UT‐ID Index thresholds for positive identification varied by dental region. Clinical trials using actual forensic cases are needed to investigate both the usefulness of subtraction radiography as well as the error rates for traditional visual identification by forensic odontologists, especially in difficult cases that contain no dental restorations.…”

Influence of observer’s experience on diagnosing simulated periapical lesions

“…In the present study, we demonstrate that experience is not a determining factor in identifying normal and pathological anatomical and periapical lesions; however, Flint et al 19 mention that experienced forensic dentists have higher level of accuracy in radiological identification than those without experience.…”

Collection and recording of radiological information for forensic purposes

“…Attempts have been made to formulate a method where similarity of two images can be mathematically expressed in this way. 23 This is a solution that indirectly attempts to address a fundamental question underlying the whole process of comparison of dentitions by any means whatever: what is the chance of two dentitions being so similar that a convincing match might occur by sheer chance?…”

“…Subtraction imaging techniques have been applied in such circumstances. 6,23 One method utilizes commonly available digital imaging programmes such as Adobe PhotoshopÒ or The Gimp. One of the radiographs is placed as a layer in a larger digital canvas, and the second is placed in a layer on top.…”