METHOD FOR AUTOMATICALLY SEGMENTING MANDIBLE USING SHAPE-CONSTRAINED INFORMATION IN CRANIOMAXILLOFACIAL CBCT IMAGE AND APPARATUS THEREOF

摘要

According to an embodiment of the present invention, a method for automatically segmenting a mandible using shape-constrained information in a craniomaxillofacial cone beam computed tomography (CBCT) image comprises the steps of: (a) globally segmenting a mandible in a craniomaxillofacial CBCT image based on an average shape obtained from a statistical shape model of a mandible multi-detector computed tomography (MDCT) image; and (b) locally improving a mandible segmenting result in consideration of brightness value characteristics and shape information for each region. The mandible can be segmented by being robust to a noise in a craniomaxillofacial CBCT image with a low brightness value contrast ratio and the segmentation accuracy can be increased in a mandible body region including a narrow region formed with a large curvature and a condyle region of which a position is different for each patient.;COPYRIGHT KIPO 2019