[Truncated abstract] Surgical simulation can extend surgeons' ability to learn, plan and carry out surgical interventions more accurately and less invasively. Despite advances in computational biomechanics, modelling of soft tissue cutting remains one of the most challenging problems in surgical simulation. The significant challenges are posed by the complexity of introducing the cutting-induced discontinuity, the difficulty of modelling geometric and material nonlinearities and the need to achieve a high computation speed. Most published works of surgical simulation use mass-spring models or isotropic linear elastic models to achieve fast computation speed. However, these models cannot account for the intrinsic nonlinear behaviour of soft tissue. In this thesis, a set of novel Meshless Total Lagrangian Adaptive Dynamic Relaxation (MTLADR) algorithms, which takes into account both geometric and material nonlinearities, was developed to robustly simulate the responses of soft tissue during surgery. These algorithms consist of the MTLADR two-dimensional (2D) algorithm, the MTLADR 2D cutting algorithm, the MTLADR three-dimensional (3D) algorithm and the MTLADR 3D cutting algorithm. Belonging to the element-free Galerkin (EFG) family, these algorithms feature a spatial discretisation and approximation based solely on nodes. The accuracy of the algorithms was verified against the established nonlinear static solution procedures available in the commercial finite element software Abaqus. The MTLADR 2D algorithm was developed for 2D simulation of large deformations of soft tissue prior to surgical cutting. The rationale for using the EFG method, the total Lagrangian formulation and the adaptive dynamic relaxation (DR) was demonstrated in this algorithm.
|Qualification||Doctor of Philosophy|
|Publication status||Unpublished - 2012|