Three-dimensional visualization of composite fillings for dental identification using CT images

Abstract

Objectives: This study aimed to discriminate between enamel and composite resins by differences in Hounsfield units shown on 16 section multidetector CT (MDCT) images taken of unidentified bodies. Methods: First, we determined the Hounsfield units of composite resins in 15 extracted human teeth. We then filled a single cavity prepared in each of the teeth with one of five different types of composite resins, and scanned the teeth using our routine post-mortem CT protocol for the head and neck. Obtained data were transferred to a radiological workstation and reconstructed. Furthermore, post-mortem CT images of the head of three unidentified bodies were reconstructed in the same manner. Results: Four types of composite resins containing radio-opaque fillers showed a constant value of 4000 HU, and one radiolucent composite resin showed values in the range of 660-800 HU in the extracted teeth. Pixels at 4000 HU indicated that the composite resins were selected and visualized as three-dimensional colour images. Composite resins could be visualized on reconstructed images of the three unidentified bodies, and the sites visualized matched those noted on the forensic dental charts. Conclusions: Discriminating enamel and composite resins containing radio-opaque materials was difficult because of their similar Hounsfield unit values. However, we did succeed in visualizing the composite resins despite limitations of the CT scale. CT reconstructed images can contribute to dental identification, particularly in cases where it is difficult to detect composite resins on external investigation, and these images can be prepared during routine dental identification work.

Investigation of an optimized scanning protocol for the dentomaxillofacial region using 320-slice multidetector computed tomography

“…In our study, subjective visualization suggested that bone crest delineations have a minimal effect on image quality, with scores ranging from 1.75 to 2.30 (fair to good), consistent with the results of a previous study by Koizumi et al 20 The visualization of dentin or enamel was worst ( Table 2), indicating that they are inherently low-contrast structures (the difference between dentin and enamel regions was ,200 HU) 21 and susceptible to changes in exposure setting, especially for tube voltage settings. The visualization of mandibular structures was slightly better than that of maxillary structures, similar to the results presented by Pittayapat et al 22 This might be because most mandibular structures are well corticated.…”

Forensic Identification of Decomposed Human Body through Comparison between Ante-Mortem and Post-Mortem CT Images of Frontal Sinuses: Case Report

“…The increasing demand for CT exams in the routine of medical services triggers a modification in the contemporary AM/PM data, which migrates from the conventional radiographs to more technological imaging modalities. The human identification process benefits from this modification achieving more realistic and reliable comparison of forensic data ( 3 , 9 ). The use of high-tech imaging techniques plays a valuable role in complex cases involving decomposed/charred bodies and skeletal remains, from which soft tissues and fingerprints are frequently damaged ( 5 , 10 ).…”

Forensic odontology radiography and imaging in disaster victim identification

“…Beam hardening or 'metal flaring' often obscure not only detail in relation to a tooth but also obscures the dentition or lack of, in the immediate area to the 'flaring'. Work in the area of CT scanning and image manipulation continues to reduce and or overcome these limitations (7,8).…”