Segmentation of 3D craniofacial images is very time consuming. Thus, the current image-based assessment takes years to quantify data from few (~10) patients. Here, we introduced an advanced technique for auto-segmentation of cone beam computed tomography (CBCT) images, which allowed for quantifying maxillary volumes of 30 impacted canine patients and 30 controls within a couple days of computer processing time. The same method was also applied to isolate individual components (e.g., tooth, maxillary process, mandibular body), which were then converted to finite element (FE) models for biomechanical analysis of tooth movement. The tooth movement was measured via clinical trials and the prediction using the individualized image-based FE theorem could answer the questions of the indeterminate biomechanical system. The novel 3D technology in particular will facilitate big data analysis and skyrocket the field of orthodontic biomechanics.
Introduce the integrate advanced machine learning techniques for auto-segmentation.
Describe the clinical finding in maxillary volume of patients with impacted canines.
Describe clinical trials and analyze indeterminate biomechanics of Stage 1 tooth movement.