Normalized cumulative percentage gray-scale distributions from digitized radiographs of dry skulls can be used to evaluate alveolar bone health. With this method, manually defined regions of interest are used to identify the area of the image to be analyzed. The purpose of this study was to determine if variations in the dimensions of the regions of interest affect the gray-scale distributions. Two approaches were used. The effect of region of interest outline variations on (1) the ability to differentiate between various sized bone chips and (2) the ability to distinguish between two groups of skulls with the alveolar bone subjectively rates as normal or diseased were investigated. Four different regions of interest, which consisted of outline and area variations to a basic region, were used. Ninety-one percent of the cumulative histograms that represented chips were sufficiently separated from nonchip plots to be distinguished. Bone health was correctly assessed with a numeric classifier for 37 of 44 quadrants, for an accuracy of 84%, a specificity of 79%, and a sensitivity of 90%. Although ROI size and shape consistency in longitudinal studies are important in radiometric analyses, small variations had minimal impact in this study.