Influence of patient motion on definition of typical cephalometric reference points in digital horizontally scanning cephalometric radiography.

Background The aim of this study was to investigate the effect of defined head-motion during x-ray exposure on the identification accuracy of typical cephalometric reference points which form the basis of treatment planning.Methods By means of a dry adult human skull and a precise motion simulation system digital Cephs are acquired while certain predefined movement patterns (shift, tilt and nodding with a motion amplitude from 5 - 50 mm) of the skull were executed. They represent the movements of children and adolescents, the main group for cephalometric radiographs.The scanning time was 9.4 s per Ceph. 10 typical landmark points of cephalometric analysis were identified by 20 observers on each Ceph twice. Using a non-motion image (Ceph0) as reference, displacement was computed as vectors relative to this image. Commonly used angles and vertical and horizontal distances were calculated.Results Both inter-rater as well as intra-rater-reproducibility were perfect. There was very little change in the vertical distance N-Me, in contrast to the horizontal distance S-N which showed a large variation. So patient motion parallel to the scanning direction of the fan-beam-detector unit, heavily influence distances parallel to this direction. The ANB angle and the Maxillo-Mandibular Plane Angle (ANS-PNS to Me-Go) only varied by about 1-2°, but large enough to greatly influence a treatment plan.Conclusions The study observed a severe influence on reference point location of motion patterns parallel to the scanning direction and also on clinically relevant distances parallel to the scanning direction. Therefore, we recommend to use a horizontal scanning direction, to minimise scanning time to a minimum, or to prefer a one-shot technique if possible. Future advancements in this field may include the integration of artificial intelligence or algorithms for the purpose of motion correction.