Precision and accuracy of low-dose CT protocols in the evaluation of skull landmarks

Abstract

It was proposed to design a series of low-dose three-dimensional (3D) CT protocols with an effective dose similar to that of the conventional radiographic series (0.011-0.032 mSv) used to assess craniofacial asymmetry. It was then aimed to assess the precision and accuracy for skull landmarks recorded ex vivo using these CT protocols. Four ultra low-dose CT protocols (5 mAs, pitch 1.375/1.75; 10 mAs, pitch 1.375/1.75) were constructed on the basis of published data, dose calculations and measurements. A high-dose CT protocol was used as a reference standard examination. The protocols were used to scan a skull immersed in water. For each protocol, two observers evaluated 17 skull landmarks on two separate occasions. 3D and two-dimensional (2D) accuracy relative to the reference standard examination and inter- and intraobserver precision was calculated. The 2D accuracy was superior for the 10 mAs protocols and the 10 mAs/1.375 pitch protocol allowed one observer to achieve an accuracy of less than 2 mm for all landmarks. The 2D interrater precision for the 10 mAs/1.375 pitch protocol was superior to the other low-dose protocols and also the high-dose protocol, achieving precision less than 2 mm for 15/17 points. The 10 mAs protocols enabled greater 2D intrarater precision than the 5 mAs protocols for both observers. Using the 10 mAs/1.375 pitch CT protocol, it was possible to achieve 2D accuracy of less than 2 mm for almost all skull landmarks with an interrater precision similar to a high-dose CT protocol. Although the estimated effective dose of 0.035 mSv was slightly greater than that of the conventional radiographic series (0.011-0.032 mSv), there was the additional benefit of a 3D data set. The 10 mAs protocols allowed accuracy and precision superior to that of the 5 mAs protocols and hence they are more likely to be clinically applicable.