Mesh sensitivity of three patient-specific bone morphing methods applied to the anybody glasgow-maastricht foot model

Scaling and morphing of dynamic musculoskeletal (MSK) models is an approach used to generate patient specific bone geometries for biomechanical analysis [1]. This work highlights mesh sensitivity of three methods applied to the Anybody Glasgow-Maastricht (GM) foot model [2],[3]. Each method uses a source and target point correspondence based approach and 12 bones including the talus, calcaneus, navicular, medial cuneiform, intermediate cuneiform, lateral cuneiform, cuboid and metatarsals 1-5 are included in the analysis. Method (A) is applied to the skin surface of the foot, method (B) is applied to bones of the foot and method (C) is a statistical shape modelling (SSM) approach. A calculation of the root mean square (RMS) error shows the SSM to have the smallest error and therefore the most representative mesh.