Automatic bone segmentation for oral surgical planning

Introduction: Evolvement of digital technology encouraged the use of virtual dental and dental surgical planning and navigated surgery. The variety of cases is significant, therefore even for an expert, it is challanging to recover a traumatic case, or to design and apply implants with high anatomical precision. Objective: High-quality image acquisition and processing are essential to the process. CT (computer tomography) or CBCT (cone-beam computer tomography) images having bad initial quality greatly decrease the performance of the processing algorithms and affect the quality of the reconstructed 3D models. 3D reconstruction using manual segmentation takes several hours of work and expertise, which significantly increases the overall cost and time of 3D CAD/CAM (computer-aided design/computer-aided manufacturing) based surgical planning and production processes. Material and method: In this paper, we introduce a bone segmentation algorithm specifically designed for surgical planning, and evaluate its benefits for the aforementioned goal. The approach relies on edge detection, mathematical morphology and basic image-processing operations. Results: Performance has been evaluated on 40 series of manual segmentation. Results (precision 90%) show that the algorithm is fast and accurate so it is applicable for surgical planning. Conclusion: Our algorithm significantly reduces the execution time of segmentation and surgical planning, therefore increases the efficiency of the overall procedure.