Validation of a metal artefact reduction algorithm in CBCT images of root filled teeth: a clinical study

• Grant number: 16/10142-5
• Support Opportunities: Scholarships abroad - Research Internship - Post-doctor
• Start date: September 23, 2016
• End date: December 22, 2016
• Field of knowledge: Health Sciences - Dentistry - Dental Radiology
• Principal Investigator: Francisco Haiter Neto
• Grantee: Karla de Faria Vasconcelos
• Supervisor: Reinhilde Jacobs
• Host Institution: Faculdade de Odontologia de Piracicaba (FOP). Universidade Estadual de Campinas (UNICAMP). Piracicaba , SP, Brazil
• Institution abroad: University of Leuven, Leuven (KU Leuven), Belgium
• Associated to the scholarship: 14/17744-5 - Effectiveness of algorithms artifacts reduction in tomographic images, BP.PD

Abstract

Image quality of Cone Beam Computed Tomography (CBCT) reconstructions is seriously compromised when metals are present in the scanned volume generating some artefacts. Shadows and streaks mask both the metal objects and the anatomical structures in the vicinity, thus prejudicing or even invalidating diagnosis. One of the possible solutions to reduce the artefacts and thus significantly enhance the diagnostic image quality, consists in using a metal artefact reduction (MAR) algorithm available in particular CBCT devices. Several MAR algorithms have been tested in the last few years but most of them are in vitro studies. However, notable controversy is observed between the results with and without MAR applications. Therefore, investigations are needed to elucidate how MAR affects the degree of the reduction effect in clinical images. In this context, we propose to investigate the effect of the MAR algorithm in CBCT images acquired from patient with root filled teeth. Images acquired with Accuitomo 170 (Morita, Kyoto, Japan; clinical reference of imaging system for endodontic purposes) and Newtom VGI evo (QR, Verona, Italy) before and after MAR application. The study will be carried out on CBCT images acquired for endodontic diagnosis. CBCT datasets will be selected from those patients, who were sent for clinical endodontic diagnoses twice with a less than 1-year interval to the Dentomaxillofacial Imaging Center, with no other dental treatment in between. An initial set of 14 patients (four males and 10 females; mean age, 54 years; range, 22-77 years) will be evaluated using a score ranging from 1 up to 5 to judge the hypodense halo (dark bands), cupping artefact (distortion) and streaks. Observers will individually evaluate the images under standard viewing conditions and the evaluation will be performed twice using a gold standard software and the original software for each CBCT device. Despite the importance of the CBCT images for endodontic field, few researches investigated the MAR performance in clinical situations, which reinforces the importance of the proposed study in evaluating the effect of a MAR algorithm with potential application in CBCT devices to improve the endodontic diagnostic process.