Virtual clinical trials of total knee replacement
All total knee replacements (TKR) must go through extensive preclinical testing to gain regulatory approval. However, current pre-clinical tests are unable to identify poor implant designs prior to clinical use. This has led to high profile failures, unnecessary patient suffering and a substantial financial burden to health system. Computational modelling can provide standardised tools to assess the efficacy and safety of new devices by exploring a range of mechanical failure scenarios across a representative population. The aim of this PhD is to build a virtual cohort of patients in which new implants can be trialled in the future. In order to achieve this, the PhD project will involve collecting detailed information about a group of knee replacement patients before and after their surgery, including 3D imaging, motion analysis and activity monitoring. This information will be used to build coupled finite element and musculoskeletal models. These models will then be used to assess the performance of the implanted knees and compared with the actual clinical outcome. The results of this project will bring the concept of virtual clinical trials closer to reality in the field of orthopaedics.
Numerically enhance physical testing of fracture fixation devices
Physical testing of the orthopaedic devices will remain an integral part of the regulatory approval process. However, these tests only examine a single failure scenario under idealised/simplified loading conditions. Simulations using validated computational models can be used to explore a wider range of loading and boundary conditions, as well as test the entire device size range, including the influence of tolerancing, and all potential combinations of parts. This is particularly relevant to fracture plating systems with variable angle screws, as the device can be configured in 100’s of different ways. Initially finite element (FE) models will be developed and validated for a range of standard regulatory tests for fracture fixation devices. These models will be used to explore the influence of variability, due to loading conditions or tolerancing, but also the number, position and orientation of the screws. FE simulations tend to be slow and are impractical when 1000’s to 10,000’s of simulations are required. Surrogate modelling techniques, however, are ideally suited to this type of problem. The technique relies on developing a black box emulation, based on an initial training set, to capture the response of the system. Computationally cheap, this can be used to explore variability, however, the performance of the surrogate model must be examined and verified. The speed, reliability and validity of surrogate modelling to enhance and expand in vitro test will be explored through this project. This project will develop enhanced in vitro and in silico test methods to better assess performance of medical devices, which will lead to better evidence generation necessary for regulatory approval and more robust devices in clinic.