Effect of a proprietary motion artifact correction algorithm on CBCT spatial resolution: an in vitro phantom study

To evaluate the effect of a proprietary motion artifact correction algorithm on CBCT spatial resolution using an in vitro phantom model.

Fifty CBCT scans were acquired with a ProMax 3D unit using a point spread function phantom mounted on a motorized turntable. Ten protocols were tested, including no motion and induced motion up to 9 degrees, with and without activation of the correction algorithm. Spatial resolution was assessed using full width at half maximum (FWHM), surface plots, fast Fourier transforms, pairwise comparisons, and Bland-Altman analysis.

Motion increased FWHM values and distorted the point spread function, with greater irregularity as motion magnitude increased. The correction algorithm produced partial improvement in selected motion protocols, particularly at lower motion magnitudes, but the effect was inconsistent across all motion conditions and measurement directions. In the no-motion protocol, activation of the algorithm introduced slight unsharpness and increased heterogeneity in the high-frequency domain.

Patient motion degrades CBCT spatial resolution. The proprietary correction algorithm may provide partial improvement under selected conditions, particularly at lower motion magnitudes, but it does not replace the need for careful patient stabilization during image acquisition and should be used cautiously when motion is unlikely.