PhD Studentship: Virtual brain networks to guide deep brain stimulation in childhood epilepsy

Around a third of children with epilepsy continue to have seizures despite medication. For those who cannot be helped by conventional surgery, electrical stimulation delivered by an implanted device is one of the few remaining options. Its use is expanding quickly, yet responses vary widely between children, and there is still no principled way to decide where to stimulate, or with what settings, for an individual patient. These choices are currently made by careful trial and error over many months.

This project uses mathematical models of brain networks to help predict efficacy of stimulation. Seizures are increasingly understood as emergent properties of brain networks, which makes them well suited to the tools of nonlinear dynamics and network science. We use multimodal imaging to develop patient-specific, mathematical models of brain networks that can be used to quantify the dynamics of brain networks. These models can then be used to predict the effect of stimulation and to probe the location and timings that most effectively alleviate seizures.

You will develop methods for constructing these models from patient recordings, implement and simulate stimulation within them, and evaluate how far their predictions correspond to measured effects using clinical data. The project therefore combines many interesting problems in mathematics, statistics and computational neuroscience including development of physiological models of neurostimulation, understanding and quantifying the dynamics of networked systems, model and parameter inference, uncertainty quantification and efficient simulation of large dynamical systems. There is scope to shape the direction of the work according to your own interests.

You will join a supportive research group at Exeter with established expertise in the mathematics of brain dynamics, and will be co-supervised by clinicians working in paediatric neurosurgery. You therefore have the opportunity to gain unique insight and training in real world clinical uses of mathematics and statistics.  We welcome applicants from mathematics, physics, engineering, computer science, quantitative neuroscience or related disciplines. Prior experience of neuroscience or clinical research is not required, and training will be provided.

Informal enquiries are warmly encouraged.

Funded by: EPSRC Engineering and Physical Sciences Research Council