A Mol1, A Balasundaram. 1. Department of Diagnostic Sciences and General Dentistry, School of Dentistry, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7450, USA. mola@dentistry.unc.edu
Abstract
OBJECTIVES: To assess the accuracy of NewTom 9000 cone beam CT (CBCT) images for the detection and quantification of periodontal bone defects in three dimensions. METHODS: A sample of 146 sites in 5 dry skulls provided the ground truth (GT). Half of the sample had bone loss of at least 3 mm. Two metal spheres at each site ensured correspondence between GT and CBCT measurements. Skulls were submerged in water and scanned with the NewTom QR-DVT-9000. A full mouth series (FMX) was obtained of each skull using photostimulable phosphor plates. Six observers measured the bone height of each site and rated the presence or absence of bone loss. Measurements were compared to GT and A(z)-values were calculated from receiver operating characteristic curves. RESULTS: The A(z)-value for CBCT was 0.74 (standard deviation (SD) = 0.14) and for FMX 0.48 (SD = 0.09). The difference was significant (ANOVA: P < 0.01). The diagnostic accuracy of CBCT was lower for anterior teeth (A(z) = 0.59) than for molars (A(z) = 0.82) and premolars (A(z) = 0.79) (Tukey's HSD (honestly significant difference): P < 0.01). The mean absolute difference between CBCT and GT was 1.27 mm (SD = 1.43) and between FMX and GT 1.49 mm (SE = 1.24) (ANOVA: P < 0.01). Measurements in the anterior mandible were less accurate than in other areas (Tukey's HSD: P < 0.01). CONCLUSION: The NewTom 9000 cone beam CT scanner provides better diagnostic and quantitative information on periodontal bone levels in three dimensions than conventional radiography. The accuracy in the anterior aspect of the jaws is limited.
OBJECTIVES: To assess the accuracy of NewTom 9000 cone beam CT (CBCT) images for the detection and quantification of periodontal bone defects in three dimensions. METHODS: A sample of 146 sites in 5 dry skulls provided the ground truth (GT). Half of the sample had bone loss of at least 3 mm. Two metal spheres at each site ensured correspondence between GT and CBCT measurements. Skulls were submerged in water and scanned with the NewTom QR-DVT-9000. A full mouth series (FMX) was obtained of each skull using photostimulable phosphor plates. Six observers measured the bone height of each site and rated the presence or absence of bone loss. Measurements were compared to GT and A(z)-values were calculated from receiver operating characteristic curves. RESULTS: The A(z)-value for CBCT was 0.74 (standard deviation (SD) = 0.14) and for FMX 0.48 (SD = 0.09). The difference was significant (ANOVA: P < 0.01). The diagnostic accuracy of CBCT was lower for anterior teeth (A(z) = 0.59) than for molars (A(z) = 0.82) and premolars (A(z) = 0.79) (Tukey's HSD (honestly significant difference): P < 0.01). The mean absolute difference between CBCT and GT was 1.27 mm (SD = 1.43) and between FMX and GT 1.49 mm (SE = 1.24) (ANOVA: P < 0.01). Measurements in the anterior mandible were less accurate than in other areas (Tukey's HSD: P < 0.01). CONCLUSION: The NewTom 9000 cone beam CT scanner provides better diagnostic and quantitative information on periodontal bone levels in three dimensions than conventional radiography. The accuracy in the anterior aspect of the jaws is limited.
Authors: Marcio Rodrigues de Almeida; Cristina Futagami; Ana Cláudia de Castro Ferreira Conti; Paula Vanessa Pedron Oltramari-Navarro; Ricardo de Lima Navarro Journal: Dental Press J Orthod Date: 2015 May-Jun