Reliability and accuracy of cone-beam computed tomography dental measurements

Abstract

Dental measurements are an integral part of the orthodontic records necessary for proper diagnosis and treatment planning. In this study, we investigated the reliability and accuracy of dental measurements made on cone-beam computed tomography (CBCT) reconstructions. Thirty human skulls were scanned with dental CBCT, and 3-dimensional reconstructions of the dentitions were generated. Ten measurements (overbite, overjet, maxillary and mandibular intermolar and intercanine widths, arch length available, and arch length required) were made directly on the dentitions of the skulls with a high-precision digital caliper and on the digital reconstructions with commercially available software. Reliability and accuracy were assessed by using intraclass correlation and paired Student t tests. A P value of < or = 0.05 was used to assign statistical significance. Both the CBCT and the caliper measurements were highly reliable (r >0.90). The CBCT measurements tended to slightly underestimate the anatomic truth. This was statistically significant only for compounded measurements. Dental measurements from CBCT volumes can be used for quantitative analysis. With the CBCT images, we found a small systematic error, which became statistically significant only when combining several measurements. An adjustment for this error allows for improved accuracy.