A new, highly precise measurement technology for the in vitro evaluation of the accuracy of digital imaging data

Three-dimensional radiological imaging data play an increasingly role in planning, simulation, and navigation in oral and maxillofacial surgery. The aim of this study was to establish a new, highly precise, in vitro measurement technology for the evaluation of the geometric accuracy down to the micrometric range of digital imaging data.

A macerated human mandible was scanned optically with an industrial, non-contact, white light scanner, and a three-dimensional (3D) model was obtained, which served as a master model. The mandible was then scanned 10 times by cone beam computed tomography (CBCT), and the generated 3D surface bone model was virtually compared with the master model. To evaluate the accuracy of the CBCT scans, the standard deviation and the intraclass coefficient were determined.

A total of 19 measurement points in 10 CBCT scans were investigated, and showed an average value of 0.2676 mm with a standard deviation of 0.0593 mm. The standard error of the mean was 0.0043 mm. The intraclass correlation coefficient (ICC) within the 10 CBCT scans was 0.9416.

This highly precise measuring technology was demonstrated to be appropriate for the evaluation of the accuracy of digital imaging data, down to the micrometric scale. This method is able to exclude human measurement errors, as the software calculates the superimposition and deviation. Thus inaccuracies caused by measurement errors can be avoided. This method provides a highly precise determination of deviations of different CBCT parameters and 3D models for surgical, navigational, and diagnostic purposes. Thus, surgical procedures and the post-operative outcomes can be precisely simulated to benefit the patient.