Quantitative evaluation of alveolar bone resorption on dental panoramic radiographs by standardized dentition image transformation and probability estimation.

Alveolar bone resorption, which can be observed on dental panoramic radiographs (DPRs), is one of the most important signs of the progression of periodontal disease. However, quantitative and consistent evaluation of alveolar bone resorption is labor intense, because of a large number of DPR examinations performed each year. In our previous study, we developed an automated scheme for detection of three lines, corresponding to an occlusion line, an alveolar line and a root apex line, to measure the alveolar bone ratio (ABR), which may be useful to dentists in early detection of periodontal disease. In this study, we propose a new image transformation that maps the tooth region to a standardized dentition image for improved visualization and detection of alveolar crest points by probability estimation. The proposed method was applied to 92 DPRs with 4 degrees of alveolar bone resorption. The ABR measurement obtained by the proposed method was compared with the manual measurement of alveolar bone loss (ABL). In this preliminary investigation, the computer-estimated ABR was moderately correlated with the manual ABL. The result indicates the potential usefulness of the method for early detection and consistent diagnosis of periodontal disease.