作者: Sylvia Rueda , José Antonio Gil , Raphaël Pichery , Mariano Alcañiz
DOI: 10.1007/11866565_21
关键词:
摘要: Preoperative planning systems are commonly used for oral implant surgery. One of the objectives is to determine if quantity and quality bone sufficient sustain an while avoiding critical anatomic structures. We aim automate segmentation jaw tissues on CT images: cortical bone, trabecular core especially mandibular canal containing dental nerve. This nerve must be avoided during surgery prevent lip numbness. Previous work in this field thresholds or filters needed manual initialization. An automated system based use Active Appearance Models (AAMs) proposed. Our contribution a completely semi-automatic landmarking process necessary create AAM model. The trained using 215 images tested with leave-4-out scheme. Results obtained show initialization error 3.25% mean 1.63mm 2.90mm core, 4.76mm 3.40mm