MASTER PROJECTS AVAILABLE IN THE GROUP
Pull-out prediction system of the osteoporotic pedicle screws
The design of the osteoporotic pedicle screw is an interesting topic for decades, while the low pull-out strength usually causes the loosening of bone-screws-interfaces. The shape optimization of pedicle screws is a simple and feasible way to improve performance. Most of the optimizations are based on numerical simulation, as it could provide objective values for different design parameters in a short period. However, the numerical simulation sometimes does not represent the real situation because of the simplifications and several assumptions when we build the geometry, material model, and boundary conditions. To overcome the problems mentioned for numerical simulation, we can use experimental data to predict the pull-out strength. The only problem with experiments is the one to one projection, which means, the experimental data can only represent the objective value for specified input. And usually, the experiments take time and sources, thus we cannot afford too many experiments. However, machine learning/statistical analysis is a solution in this field, which is used for predictions based on big data.
In this project, we want to build a prediction system for osteoporotic screws, which is based on the experimental data and the machine learning statistical analysis model. We use sawbones and several designed screws from DOE to do the pull-out tests. The outputted pull-out strength and stiffness could be utilized to create the response surface of different screw-designs. Different machine learning/statistical analysis models will be implemented or inputted from Python or MATLAB libraries. The prediction results can be verified and validated with extra pull-out experiments.
In vitro cell response to ions and particles released from spinal implant materials
Spinal implants (e.g total disc replacement, spinal cages or spinal instrumentation such as pedicle screw and rod systems) are medical devices used to relieve pain, stabilise and treat atypical spinal curvature. These devices have been used for several decades, however the population is ageing and more active and younger patients undergo spinal surgery, putting higher demands on their performance. For that reason, prolonging the lifetime of spinal implants is of great interest and improving the resistance to corrosion and wear of implant materials is an important area of study. Nevertheless, particles and ions will inevitably be generated to some extent due to implant degradation and will be released to the periprosthetic tissue. It is well documented that implant debris triggers local immune and inflammatory responses, which worst case can lead to revision surgery. Spinal implants are of particular concern due to their proximity to the spinal cord and nerves. Recent studies have shown evidence of disruption of the meninges by these particles compromising their barrier effect, leading to the migration of particles and ions into the spinal cord. To date most of the biocompatibility studies on wear debris and ions have been done using cells from other tissues, and the effects of particles and ions on neural cells remain unclear.
In this project, we are developing an advanced 3D in vitro model of the spinal cord using neural cells to evaluate their biological response to wear and corrosion products from spinal implant materials. The gold standard in in vitro testing is the use of 2D cell culture, so the comparison with the 2D culture system is considered relevant, and would be the main topic of the MSc thesis project. Different laboratory techniques (e.g fluorescent staining, ICC, ELISA, colourimetric/luminescent techniques) will be considered to measure viability, morphology, and inflammatory response of the cells after exposure to the particles/ions.
Please send your CV and motivation letter to Estefanía Echeverri Correa: firstname.lastname@example.org