Processing of brain time series to unravel changes in dynamic functional connectivity in disease

This exciting inter-disciplinary PhD project will develop cutting-edge computational and mathematical analysis methods to monitor and understand the effects of major neurological diseases.

The successful applicant will be trained in an inter-disciplinary and emerging area of research at the interface of signal processing and network theory, in collaboration with renowned clinical collaborators and benefitting from datasets acquired in real-world clinical settings. The PhD will prepare the successful candidate for post-doctoral, more independent research. Applications of signal processing and network theory are expanding nowadays, and this PhD provides an excellent opportunity to be trained in these areas.

We are motivated by the fact that neurological diseases are one of the greatest threats to public health. We need new tools to tackle this looming crisis in the form of better personalized models of disease. We will focus on processing electroencephalogram (EEG) signals, which record brain activity directly and non-invasively at a number of locations over the scalp. These multivariate time series offer unmatched opportunities to assess brain activity.

We have recently recognized the importance of brain connectivity for healthy brain function. This refers to the concept that different parts of the brain need to interact with each other in a coordinated way. In the EEG, this is assessed through the analysis of functional connectivity: statistical dependencies between multivariate time-varying recordings of brain activity acquired at distinct locations. Network science is then used to study the connectivity patterns from a system’s perspective by decomposing them into a set of elements and relationships that link the elements. However, despite the relevance of these approaches, their actual practical use in the clinic is still limited. This is probably because current methods tend to disregard the dynamical nature of brain activity that makes connectivity patterns evolve rapidly in time. Thus, we need new signal processing and network analyses to better characterize the dynamical and multifaceted nature of brain activity.

Therefore, the main objective of this inter-disciplinary project is to create novel signal processing methods to reveal rich temporal and topological features currently disregarded in the connectivity analyses of multivariate EEG time series and to assess their usefulness for the monitoring of major neurological conditions such as epilepsy or dementia.

Further Information: 

Early application is encouraged.

Closing Date: 

Sunday, February 24, 2019
Engineering School Logo
Engineering School Logo

Principal Supervisor: 


Enthusiastic and self-motivated candidates are sought with a background in electronic engineering, computer science, mathematics or cognate disciplines. Candidates should have a Master’s or, in exceptional cases, an excellent Bachelor’s degree (2:1 or higher). Previous experience in areas related to graph theory (e.g., networks, algebra, etc.) or signal processing (e.g., time series analysis) would be beneficial but it is not necessary. The candidate is expected to have good programming and analytical skills.

Further information on English language requirements for EU/Overseas applicants.


We welcome applications from UK and EU students eligible for Research Council funding, and from strong students from other nationalities interested in applying to scholarships from the University of Edinburgh or elsewhere, or who may have their own funding arrangements.

Further information and other funding options.

Informal Enquiries: