The Engineering and Physical Sciences Research Council (EPSRC) has today announced £9m of funding – known as the Healthcare Technologies Challenge Awards – to be shared among research projects that promise to improve healthcare diagnosis and treatment across a wide range of issues. Life Sciences Minister George Freeman described the projects as 'innovative' and 'game-changing', illustrating why the UK is a world leader in life sciences.
Professor Antoine Jérusalem: 'Electrophysiological-mechanical coupled pulses in neural membranes: a new paradigm for clinical therapy of SCI and TBI (NeuroPulse)'
'Traumatic brain injury (TBI) is a major public health issue, with over 10 million incidents resulting in death or hospitalisation globally each year. The World Health Organization estimates that TBI will surpass many diseases as the major cause of death and disability by 2020. The other constituent of the central nervous system is the spinal cord. Damage to the spinal cord, or spinal cord injury (SCI), shares many similarities with TBI but can lead independently to pain and/or motor impairments ranging from incontinence to paralysis. Pain management – a direct consequence of TBI and SCI but much wider in scope – is quietly emerging as one of the most important healthcare costs in the UK.
'The recent increase in the number of research campaigns on TBI and SCI has drastically improved the understanding of the coupled role of micromechanics and electrophysiology at the neuron level. However, the intrinsic relationship between mechanical vibrations and their biophysical implications is still widely ignored in this context. As a direct consequence, the effect of functional alteration due to a mechanical insult on the vibrational mechanical properties of a neuron has so far been ignored.
'NeuroPulse thus aims at developing and utilising state-of-the-art modelling approaches for the study of electrophysiological and mechanical coupling in a healthy and mechanically damaged axon, nerve and, eventually, spinal cord and brain white-matter tract. The resulting in-silico platform will be calibrated and validated by means of a comprehensive experimental programme in collaboration with the Department of Physics at Oxford, via co-investigator Professor Sonia Contera. Two teams of clinical project partners in Oxford and Cambridge will participate in the analysis of the results for direct applications in a clinical setting. The clinical support for the research at Oxford will be provided by Dr Damian Jenkins and Mr Tim Lawrence of Oxford University Hospitals NHS Foundation Trust.
'NeuroPulse encompasses three of the four pillars of the Healthcare Technologies Grand Challenges: a) the development of future therapies aimed at leveraging neuron electrophysiological-mechanical coupling to modify or quantify electrophysiological properties through control of the mechanical properties; b) treatment optimisation by non-invasive mechanically based pain treatment and functional deficit detection; and c) transforming community health and care by alleviating the need for primary care to "second guess" a patient's symptoms (as is unfortunately often the case in TBI), and by providing an objective roadmap for damage quantification.
'In doing so, NeuroPulse will help decrease the societal and economic burden of TBI, SCI and pain management in the UK and worldwide.'
Dr David Clifton: 'Machine Learning for Patient-Specific, Predictive Healthcare Technologies via Intelligent Electronic Health Records'
'Healthcare systems worldwide are struggling to cope with the demands of ever-increasing populations in the 21st century, where the effects of increased life expectancy and the demands of modern lifestyles have created an unsustainable social and financial burden. However, healthcare is also entering a new, exciting phase that promises the change required to meet these challenges: ever-increasing quantities of complex data concerning all aspects of patient care are being routinely acquired and stored throughout the life of a patient. These datasets include the electronic health records (EHRs) now active in many hospitals, as well as data from patient-worn sensors.
'The dramatic growth in data quantities far outpaces the capability of clinical experts to cope, resulting in a so-called "data deluge" in which the data are largely unexploited. There is huge potential for using advances in big-data machine learning to exploit the contents of these complex datasets by performing robust, automated inference at very large scales. This promises to improve healthcare outcomes significantly by allowing the development of patient-specific healthcare technologies, a field in which there is little existing research.
'The Oxford project, based in the Computational Health Informatics Laboratory and collaborating closely with the Oxford University Hospitals NHS Foundation Trust, will tackle two main problems.
'First, there is a global need for reliable, continuous monitoring of patients in healthcare systems, both in hospitals and hospital-at-home settings. The majority of patients, though, are ambulatory and unmonitored, and, consequently, there is an unacceptably high rate of mortality and morbidity. However, existing monitoring systems are simplistic, have very high false alarm rates, and are not predictive. We will develop a suite of technologies that exploit the very large quantities of largely unused data, enabling clinicians to have early warning of patient deterioration via monitoring systems that forecast patient health in a robust manner, using machine-learning methods.
'Second is the idea that infectious disease is as great a threat to national security as climate change. Existing lab-based methods for testing for infectious disease can take up to six weeks to perform. However, whole-genome analysis (of bacterial DNA) combined with machine learning across the EHR could be performed almost on the same day, greatly improving our ability to identify and fight outbreaks of infectious disease. Here, the engineering challenge is to develop probabilistic models to identify novel relationships between the massively multivariate genomic data of the pathogen (the genotype) and the corresponding patient data from the EHR (the phenotype). In this second theme, we will improve our understanding of infectious disease, and, working with Public Health England, will provide systems that predict resistance to antibiotic drugs so that patients are given the right drugs at the right time. The changing nature of drug resistance is a serious problem for the way we deliver healthcare, and machine-learning systems that can "evolve" their understanding of resistance through time are of particular importance.'