Computer Vision Support for the Orthodontic Diagnosis

Abstract

The following paper presents the achievement reached by our joined teams: Computer Vision System Group (ZKSW) in the Institute of Theoretical and Applied Informatics, Polish Academy of Sciences and Department of Orthodontics, Silesian Medical University. The cooperation began from the inspiration of late Prof. A. Mrozek. Computer Vision in supporting orthodontic diagnosis means all the problems connected with proper acquisition, calibration and analysis of the diagnostic images of orthodontic patients. The aim of traditional cephalometric analysis is the quantitative confirmation of skeletal and/or soft tissue abnormalities on single images, assessment of the treatment plan, long term follow up of growth and treatment results. Beginning with the computerization of the methods used in traditional manual diagnosis in the simplest X-ray films of the patient's head we have developed our research towards engaging different methods of morphometrics, deformation analysis and using different imaging modalities: pairs of cephalograms (lateral an frontal), CT-scans, laser scans of dental models, laser scans of soft tissues, finally merging all the image information into patient's specific geometric and deformable model of the head. The model can be further exploited in the supporting of the surgical correction of jaw discrepancies. Our laboratory equipment allows us to design virtual operations, educational programs in a virtual reality with a CyberGlove device, and finally to verify the plan of intervention on stereo lithographic solid models received from a 3D printer.