Quantitative Evaluation of Proximal Bone Lesions using Digital Subtraction Radiography

Digital subtraction of radiographs has been demonstrated to be sensitive for detecting alveolar crestal bone changes and may be used for their quantitative analysis. The purpose of this study was to evaluate the subtraction method developed here in terms of lesion extent and average optical density in the region of interest. A lesion of increasing size was produced in the crest between the first and second molars in seven human dry mandibles. Standardized radiographs were taken at each step using three different exposures. These images were digitized using a drum-scanning microdensitometer coupled to an image processor. Differences in film contrast were corrected by comparison of corresponding gray levels between pairs of images. Subtraction images were produced from successive projections with and without lesions. The results indicate that digital subtraction together with contrast correction made possible detection and measurement of minute changes of diagnostic interest.