Influence of Artifact Reduction Tools in Micro-computed Tomography Images for Endodontic Research.

INTRODUCTION: Micro-computed tomography (μCT) is an imaging modality of growing application in endodontic research because of its nondestructive technology that enables visualization at the micrometer level. In the presence of high-density material, images acquired with μCT may present artifacts. Thus, the aim of this study was to evaluate subjectively and objectively the influence of artifact reduction tools available for μCT image reconstruction to determine whether their use can influence endodontic research outcomes.

METHODS: Ten teeth were scanned in a μCT device, and these images were reconstructed by applying 13 protocols of artifact reduction tools, combining ring artifact reduction (RAR) and beam-hardening artifact reduction (BAR). Images were assessed subjectively (observer's preference) and objectively (root canal surface area and volume). Observer's preference for RAR and BAR protocols was tested by χ2 . Analysis of variance was used to compare volumes and surface area of root canals on different protocols. Intraobserver and interobserver reproducibility was calculated by the weighted kappa test.

RESULTS: There was no preference for a particular protocol for BAR (P = .91) or RAR (P = .80). There were no statistically significant differences for volume (P = .999) or surface area (P = .972) of root canals for all protocols.

CONCLUSIONS: To evaluate root canal volume and root canal surface area, artifact reduction tools applied to μCT images can be used according to the observer's visual preference without influence on objective image analysis.