Virtual Anatomical Reconstruction of Mandibular Bone Defects Using a Statistical Shape Model

Virtual Anatomical Reconstruction of Mandibular Bone Defects Using a Statistical Shape Model

Hundred already taken CT-scans (50 male and 50 female) will be anonymized. These scans will be segmented and implemented into a statistical shape model that will be used to predict missing hard tissue of a mandible based on the remaining bone.

Together with the radiology and Craniomaxillofacial (CMF) department, 100 scans will be selected from the database (50 males and 50 females, between 20 and 60 years old). First, they will be anonymized (except the sex and age/date of birth). Afterwards, they will be segmented using Materialise Mimics software. The segmented mandibles will then be used to create a statistical shape model (SSM) using principal component analysis. This model will afterwards be used to predict the mandibular shapes.

Inclusion Criteria:

• Healthy mandibles

Exclusion Criteria:

• Mandibles with major deformities
• Mandibles with obvious problems
• Mandibles that already underwent surgery