EPSRC Centre for Predictive Modelling in Healthcare First Seed Corn Projects Kick Off

The EPSRC Centre for Predictive Modelling in Healthcare commenced its first seed corn projects in October 2016.

The Centre’s seed corn projects are multidisciplinary research projects that bring together researchers with complimentary expertise, to work on a research problem associated with the expertise of researchers within Centre. This round of seed corn funding was provided by King’s College London and up to £10k could be requested for each project.

The projects began with a seed corn incubator event held on 10-11 October in Ascot, which brought together the researchers at both the University of Exeter and Kings College London to finalise their project plans and set themselves up ready for the projects to begin.  Four seed corn projects were funded and a summary of each is provided below:

A patient specific statistical-systems approach to mechanistically stratify heat failure with preserved ejection fraction patients based on the cause of increased ventricular stiffness

Heart failure with preserved ejection fraction is a prevalent and progressive disease affecting around 450,000 patients in the UK. To date, not a single drug has been discovered to improve the life expectancy of people touched by this illness. To overcome this situation, we have created a team of experts from different backgrounds, Dr Steve Niederer and Dr Sander Land, experts in cardiac modelling from King’s College London, and Professor Peter Challenor and Lauric Ferrat, experts in statistics from the University of Exeter. This seed corn project involves bringing together two strands of mathematics; cardiac modelling and uncertainty quantification in order to better understand the heart. This combination of skills and mathematics will lead to a deeper understanding of the causes of the illness and could ultimately lead to better treatments for diastolic heart failure.

The hypothalamic KNDy neural network underlies the emergence of the GnRH pulse generation

The ability to reproduce is fundamental for survival and is controlled by the reproductive system that coordinates activity in the brain, the pituitary gland and the ovaries in women and the testes in men.  Within the brain is a clock or neural oscillator that drives the release of intermittent pulses of a brain hormone called gonadotrophin-releasing hormone (GnRH), which sets in motion a cascade of hormonal signals that lead to stimulation of the ovaries and testes. The nature of this molecular clock has remained elusive, but about 10 years ago a new set of brain cells were discovered that contained 3 particular brain chemicals, or neuropeptides: two well-known ones called Neurokinin-B and Dynorphin; and a totally new one called Kisspeptin. These new cells are called KNDy neurons, and are thought to form the network that drives the periodic release of GnRH. This seed corn project will seek to understand how the KNDy neurons operate to generate GnRH pulses, using a combination of experimental manipulation and mathematical modelling. The project involves Professor Kevin O’Byrne and Dr Xiao Feng Li from King’s College London and Professor Krasimira Tsaneva-Atanasova and Dr Margaritis Voliotis from the University of Exeter. The team’s approach offers a unique opportunity to discover the nature of the GnRH pulse generator, which is at the centre of reproduction, and reveal its operational characteristics. This will help future developments of more effective treatments for fertility disorders and personalised medicines in reproductive health.

Developing mathematical methods to infer mental health states from wearable technology

Many mental health conditions are enduring and likely to affect people for prolonged time periods. Service users, carers and the healthcare system often have to deal with crisis and relapse episodes associated with symptoms worsening. There is increasing recognition in the field that regular monitoring may be particularly important to achieve early detection of emerging problems and prevention of relapse. However, mental healthcare services currently face several challenges, including the limited availability of tools for preventing hospitalisation and the regular monitoring of patients. There is hope that mobile technology could help assess health states regularly, in real-time, remotely, and cost-effectively. However, the information that wearable devices currently provide is complex and does not correspond directly to specific indicators of mental health.

This seed corn project brings together psychologist Dr Matteo Cella and biostatistician Dr Daniel Stahl from King’s College London and mathematicians Dr Jamie Walker and Dr Eder Zavala from the University of Exeter. The team aims to develop a tool that reliably predicts emotional states associated with mental health conditions such as anxiety, low mood, and stress by analysing data collected by wearable devices. This has the potential to contribute to fulfilling a long-sought goal of empowering patients to monitor their own health conditions, improving mental health outcomes that reduce the societal costs associated with mental illness (e.g. loss of employment, welfare costs), and optimising the use of healthcare resources by preventing symptoms from becoming severe.

Computer-aided diagnosis and prognosis of epilepsy

The group has recently demonstrated that mathematical analysis and modelling of routinely acquired clinical EEG data can classify a subject with suspected epilepsy with high accuracy. Building on this research, this seed corn project seeks to develop a prototype decision support tool for epilepsy. To achieve this, Scientific Programmer Dr Diane Fraser is working with Dr Wessel Woldman and Professor John Terry at the University of Exeter, in collaboration with a group of clinical researchers led by Professor Mark Richardson from King’s College London. Conversations with a variety of clinicians confirmed the potential of these methods as the basis of a decision support tool, where several envisioned these methods could support clinical practice by revealing valuable additional information in a fast, objective and accurate fashion. Crucially, a number of neurologists have expressed their interest in testing the prototype in their clinical settings once finished. By integrating the feedback from potential early adapters, the team aims to develop a user-friendly working prototype that optimally meets clinical demands at the end of this project.

The next round of seed corn funding will commence in June 2017, please check online for details.

Leave a Reply