Interactive Segmentation and Three-Dimension Reconstruction for Cone-Beam Computed-Tomography Images

We develop a novel interactive segmentation and 3-D visualization system for cone-beam computedtomography (CBCT) data in a personal computer environment. Its design includes both a new user interface to ease the interactive manual segmentation of the volume rendered and artifact-suppressing algorithms to improve the quality of CBCT images. Four pipelines of noise removal were carried out by applying threshold filter algorithm only and with additional smoothing algorithms (Gaussian filter, Median filter, Dilation and Erosion filter). The interactive intensity plot was implemented to allow user to manually identify each anatomic structure and view the instant feedback. The visual feedback necessary during 3-D segmentation was provided by a ray casting algorithm. Both orthodontic and non-orthodontic patients demonstrated that the system was able to minimize the artifacts caused by CBCT and metal appliances. Based on the feedback obtained from the dental experts, the interactive segmentation and 3-D reconstruction functions are found to be very helpful.