Automatic detection of inferior alveolar nerve canals on CT images

Identification of the inferior alveolar nerve in the mandible is extremely important in dental implantology. With computed tomography (CT), a series of panoramic images can be generated by reformatting the three-dimensional axial data so that the alveolar nerve canals are visible in the tabular-like form on the panoramic images. Due to a rather complicated structure of the alveolar nerve, existing softwares let a user manually trace the nerve canals. In this paper, we present a new technique for automatic detection of the alveolar nerve canals on panoramic CT images. The proposed method is based on the distance transform, the morphological operations and the characteristics of the canal shape. The results show that our approach has a potential to detect the alveolar nerve canals successfully.