Computational anatomy techniques for orthopaedic research

Using computational anatomy to create models aligning with clinical know-how and its application to the design of mandibular implants.

Computational anatomy enables analysis of biological variability throughout a population. Using statistical mathematical techniques, models can be built to represent the typical shape of an anatomical structure and its predominant patterns of variability across a given population.

During 2012, we have used these techniques to propose population-based implant design for mandibular plates, where bone shape and bone mineral density are considered into a single optimization process. Additionally, we have extended our developments on models that are able to capture the variability of bone shapes using shape descriptors directly related to the clinical application (e.g. mandible fracture) and the existing knowledge of the clinicians about the anatomy (i.e. mandible regions). This facilitates the conjunction of mathematical constructs and the clinical problem at hand.