Tooth model reconstruction based upon data fusion for orthodontic treatment simulation

This paper proposes a full tooth reconstruction method by integrating 3D scanner data and computed tomography (CT) image sets. In traditional dental treatment, plaster models are used to record patient׳s oral information and assist dentists for diagnoses. However, plaster models only save surface information, and are therefore unable to provide further information for clinical treatment. With the rapid development of medical imaging technology, computed tomography images have become very popular in dental treatment. Computed tomography images with complete internal information can assist the clinical diagnosis for dental implants or orthodontic treatment, and a digital dental model can be used to simulate and predict results before treatment. However, a method of producing a high quality and precise dental model has yet to be developed. To this end, this paper presents a tooth reconstruction method based on the data fusion concept via integrating external scanned data and CT-based medical images. First, a plaster model is digitized with a 3D scanner. Then, each crown can be separated from the base according to the characteristics of tooth. CT images must be processed for feature enhancement and noise reduction, and to define the tooth axis direction which will be used for root slicing. The outline of each slice of dental root can then be determined by the level set algorithm, and converted to point cloud data. Finally, the crown and root data can be registered by the iterative closest point (ICP) algorithm. With this information, a complete digital dental model can be reconstructed by the Delaunay-based region-growing (DBRG) algorithm. The main contribution of this paper is to reconstruct a high quality customized dental model with root information that can offer significant help to the planning of dental implant and orthodontic treatment.