An automatic panoramic image reconstruction scheme from dental computed tomography images

Panoramic images of the jaws are extensively used for dental examinations and/or surgical planning because they provide a general overview of the patient's maxillary and mandibular regions. Panoramic images are two-dimensional projections of three-dimensional (3D) objects. Therefore, it should be possible to reconstruct them from 3D radiographic representations of the jaws, produced by CBCT scanning, obviating the need for additional exposure to X-rays, should there be a need of panoramic views. The aim of this article is to present an automated method for reconstructing panoramic dental images from CBCT data.

The proposed methodology consists of a series of sequential processing stages for detecting a fitting dental arch which is used for projecting the 3D information of the CBCT data to the two-dimensional plane of the panoramic image. The detection is based on a template polynomial which is constructed from a training data set.

A total of 42 CBCT data sets of real clinical pre-operative and post-operative representations from 21 patients were used. Eight data sets were used for training the system and the rest for testing.

The proposed methodology was successfully applied to CBCT data sets, producing corresponding panoramic images, suitable for examining pre-operatively and post-operatively the patients' maxillary and mandibular regions.