Spatial resolution and angular alignment tolerance in radiometric analysis of alveolar bone change.

Abstract

This pilot study was undertaken to determine the effect of x-ray beam alignment and spatial resolution on quantification of alveolar bone using radiometric techniques. Six (6) dry mandibles were radiographed at 70 kVp, 10 mA, 0.6 seconds using D-speed film, with a bone chip (2.64, 4.10, or 6.07 mg) present or absent at 7 x-ray beam alignments (0 degree, 2 degrees horizontal, 2 degrees vertical, 4 degrees horizontal, 4 degrees vertical, 6 degrees horizontal, 6 degrees vertical). This resulted in 28 radiographs per mandible. Radiographs were digitized using 50- and 200-microns pixel spatial resolution. Image gray levels were standardized using a simple look-up table shift. Regions of interest (ROIs) were positioned on the alveolar bone where the bone chips had been placed. Cumulative percent histograms (CPH) were calculated for those ROIs. Regression analysis was used to evaluate the relationships between CPH changes and bone chip size as x-ray beam angulation and spatial resolution was varied. The resulting R2 values for angulation ranges of 0 degree to 1.4 degrees, 1.5 degrees, to 2.4 degrees, and 2.5 degrees to 5.5 degrees were: 0.983, 0.941, 0.891 for 50-microns pixel images and 0.869, 0.909, and 0.774 for 200-microns pixel images. We conclude that 50-microns pixel spatial resolution is apparently superior to 200-microns pixel images if radiometric data is to be evaluated. With 50-microns pixel spatial resolution, alignment variations up to 5 degrees may be acceptable in clinical studies, depending on the magnitude of bone change that is to be detected.