Automatic method of analysis of OCT images in the assessment of the tooth enamel surface after orthodontic treatment with fixed braces

Introduction: Fixed orthodontic appliances, despite years of research and development, still raise a lot of controversy because of its potentially destructive influence on enamel. Therefore, it is necessary to quantitatively assess the condition and therein the thickness of tooth enamel in order to select the appropriate orthodontic bonding and debonding methodology as well as to assess the quality of enamel after treatment and clean-up procedure in order to choose the most advantageous course of treatment. One of the assessment methods is computed tomography where the measurement of enamel thickness and the 3D reconstruction of image sequences can be performed fully automatically. Material and method: OCT images of 180 teeth were obtained from the Topcon 3D OCT-2000 camera. The images were obtained in vitro by performing sequentially 7 stages of treatment on all the teeth: before any interference into enamel, polishing with orthodontic paste, etching and application of a bonding system, orthodontic bracket bonding, orthodontic bracket removal, cleaning off adhesive residue. A dedicated method for the analysis and processing of images involving median filtering, mathematical morphology, binarization, polynomial approximation and the active contour method has been proposed. Results: The obtained results enable automatic measurement of tooth enamel thickness in 5 seconds using the Core i5 CPU M460 @ 2.5GHz 4GB RAM. For one patient, the proposed method of analysis confirms enamel thickness loss of 80 mu m (from 730 +/- 165 mu m to 650 +/- 129 mu m) after polishing with paste, enamel thickness loss of 435 mu m (from 730 +/- 165 mu m to 295 +/- 55 mu m) after etching and bonding resin application, growth of a layer having a thickness of 265 mu m (from 295 +/- 55 mu m to 560 +/- 98 mu m after etching) which is the adhesive system. After removing an orthodontic bracket, the adhesive residue was 105 mu m and after cleaning it off, the enamel thickness was 605 mu m. The enamel thickness before and after the whole treatment decreased by about 125 mu m. Conclusions: This paper presents an automatic quantitative method for the assessment of tooth enamel thickness. This method has proven to be an effective diagnostic tool that allows evaluation of the surface and cross section of tooth enamel after orthodontic treatment with fixed thin-arched braces and proper selection of the methodology and course of treatment.