Artificial Intelligence (AI) has advanced rapidly over the last five years, largely as a result of new algorithms, affordable hardware, and huge increases in the availability of data in digital form. The UK has recognised as a national priority the urgent need to exploit AI in human health, where digital data is being created from many sources, for example: images from tissue slices, X-ray devices, and ultrasound; along with laboratory tests, genetic profiles, and the health records used by GPs and hospitals. The potential is enormous. In future, AI could automatically identify those at risk of cancer before symptoms appear, suggesting changes in lifestyle that would reduce long-term risk. It could greatly speed-up and increase the reliability of diagnostic services such as pathology and radiology. It could help doctors and patients select the most appropriate care pathway based on personal history and clinical need. Such improvements will lead to better care and more cost-effective use of resources in the NHS. Our Centre for Doctoral Training will train the future researchers who will lead on this transformation. They will come from a variety of backgrounds in science, engineering and health disciplines. When they graduate from the Centre after four years, they will have the AI knowledge and skills, coupled with real-world experience in the health sector, to unlock the immense potential of AI within the health domain. Our scope is on AI for medical diagnosis and care with a focus on cancer for which there are particularly rich sources of digital data, and where AI is expected to lead to significant breakthroughs. Leading with cancer, we will inform the use of AI in medical diagnosis and care more widely. The Centre will be based in the City of Leeds, which has developed into the home of the NHS in England. The University of Leeds and the Leeds Teaching Hospitals Trust (LTHT), working with key national partners from the NHS and industry, provides the ideal environment for this Centre. There is internationally excellent research on AI and on cancer, including a world leading centre for digital pathology. There is already strong collaboration between the different organisations involved. The Centre builds on a well-established track record in transferring research ideas into world-leading clinical practice and new products. Our graduates will become international leaders in academia and industry, ensuring the UK remains at the forefront in health research, clinical practice and commercial innovation.

The Transmissible Spongiform Encephalopathy (TSE) diseases are a group of fatal neurodegenerative diseases which include scrapie in sheep, BSE in cattle and CJD in humans. TSE diseases (also known as prion diseases) differ from other neurodegenerative disorders such as Alzheimer's disease, due to their infectious nature. Instead of an conventional infectious agent such as a bacterium or virus the TSE infectious agent (the prion) is thought to be a misfolded form of a host protein (PrP). It has been hypothesised that the abnormally folded form of the protein (PrPSc) is able to bind to the normal protein which is found in brain tissue of all mammals, and convert it into the abnormal form. PrPSc accumulates as disease progresses, and may cause the death of neurons in the brain. PrPSc is usually found in infected tissue, and can be identified microscopically by the presence of abnormal protein aggregates in sections of brain tissue, or by its resistance to digestion with proteases on immunoblotting. PrPSc co-purifies with the TSE infectious agent, and correlates with the level of infectivity present. PrPSc was therefore thought to be the sole component of the 'prion', and is currently the only diagnostic marker used for TSE disease testing. PrPSc can exist as either diffuse deposits or large amyloid aggregates in tissue, but the role of each form in disease is unknown. Conflicting studies have suggested both an infectious and a protective role for PrP amyloid in TSE disease. In addition, other experiments have shown that PrPSc is not always present in infectious tissue. These findings raise serious questions about the suitability of PrPSc as the only available diagnostic marker, and it is important for both accurate disease diagnosis and the development of new therapies and treatments for these currently incurable diseases that we identify exactly which form of PrP is associated with infectivity. In this proposal, we aim study the amyloid form of PrPSc and its association with the infectious agent. In our laboratory we have observed that transgenic mice inoculated with brain material from a case of atypical human prion disease do not develop clinical or pathological signs of disease, but do produce large amyloid aggregates in the brain. We have been unable to transmit disease from brain tissue of mice possessing these aggregates, indicating the absence of TSE infectious agent in these tissues. Current diagnostic methods would have identified these mice as TSE infected, yet we have shown the mice lack both disease and infectious agent. Our results support the hypothesis that PrP amyloid is not infectious, and may be formed by seeding from amyloid in the inoculum, or may be a host protective mechanism by which smaller more infectious aggregates are sequestered into an inert form. We therefore aim to identify the role of amyloid in TSE disease by inoculating transgenic mice with oligomeric and amyloid forms of recombinant PrP to determine whether we can induce amyloid formation in transgenic mice in the absence of infected tissue inoculum, and whether such amyloid forms of PrP are infectious. We also aim to disrupt these amyloid deposits to determine whether smaller fragments from the amyloid are infectious. The results from these experiments will aid in our understanding of the role of PrP amyloid in TSE disease. If amyloid is a protective mechanism by which the host controls TSE infectivity, treatments which target the disruption of such aggregates may instead enhance disease, and would therefore be undesirable. These results will also help to identify specific forms of PrP associated with TSE infectivity, leading to the development of accurate diagnostic tests with low risk of both false negative and false positive results which is important ethically when developing diagnostic assays for human prion disease.

The aim of "systems biology" is to understand how biological systems (e.g. cells, organs, people) work as a collection of parts by using mathematical modelling. We describe the behaviour of the parts in the form of mathematical equations using the laws of physics and chemistry, and then see how the behaviour of larger systems emerges from this. Many systems biology models for specific components have been published, but there remain significant challenges in exploiting them to understand systems as a whole. Which existing models (if any) are most appropriate for a new scientific question? How does each model behave in different situations? How well do they capture what the real systems do? At present it is very difficult to answer these questions without first downloading each model, writing programs to perform different simulated experiments, and then writing more code to compare and visualize the results. There has been nowhere to look up even simple properties for different models. We propose to build upon a pilot implementation of a system that enables such tasks to be done automatically, with results published on a website. Our approach will be demonstrated in perhaps the most mature area of systems biology: the electrical activity of heart cells, for which the first model was published in 1960 and well over 100 models are now available in public databases. The models have been hugely important in giving insights into how the heart works (and what can go wrong due to disease, age or drug side effects) and have helped in developing new treatments. The first step is to compare the different model predictions to measurements from real cells, to tell us whether we really understand how the heart's cells work. We will link real measurements to our recipes for performing equivalent experiments on the computer models, and provide an interface to display the results, indicating how well they agree both qualitatively (general appearance) and quantitatively (how well the numbers match). Cardiac models often have dozens of equations containing hundreds of parameters - key numbers governing how the models behave. How these parameters were worked out from experimental recordings (data) is, more often than not, unclear. Since many models reuse components from previous models, the original methods and data may no longer be available to anyone. This causes big problems for building on these models - if for instance we want to adapt a model to a new cell type, there is no record of which experiments were performed, or how these were analysed to produce the parameters and equations in the final model. We will extend our recipes to capture this information as well, and so be able automatically to re-calibrate a model to a given set of experiments. Crucially, our tools will use the variability in experimental measurements to calculate how models are likely to need to change to capture variations between different cells in a heart, or between different people, and explain how this variation affects predictions. Three case studies will drive development, looking at different kinds of model to give a broad picture of needs. Feedback from the wider community and an external advisory board of experts in cardiac electrophysiology will also be incorporated, building on the success of our first user workshop in September 2015. The final output will be a user-friendly online system - a "Cardiac Electrophysiology Web Lab" - providing an open community resource for researchers to use. We will also write training materials and run further workshops to help these researchers use it. Our resource will make it easier to reuse or extend existing models in appropriate ways, to develop new models, and to understand differences between heart cells. The tools will increase the impact of modelling for replacing animal experimentation and testing, e.g. in drug trials.

Functional coatings are highly engineered drug delivery systems whose structure and composition is critical to the controlled release of the active pharmaceutical ingredient in the human body. This increase in manufacturing complexity coincides with a time when companies are looking to reduce costs while regulators exert pressure on the sector to ascertain a greater understanding of the products' critical quality attributes (CQAs) and associated process control. To date the development and manufacture of these high value products is challenging owning to the fact that pharmaceutical processing is complex and dominated by empirical knowledge with large gaps remaining in the full scientific understanding of the underlying processes. It is an essential need, and also a big business opportunity, to develop a step change technology-a "smart factory" capable of manufacturing these high-value products to user-defined specifications. This EPSRC call provides the consortium with the necessary funding to develop the basic components of a "smart factory" by the integration of process modeling and in-process diagnostic capability for real-time in-situ process control of advanced tablet manufacturing. By utilizing the unique diagnostic information obtained by a range of in-line sensors including our THz imaging and optical coherence tomography (OCT) sensors, we will develop theoretical models to identify key process parameters that will ultimately allow the development of an active feedback loop for advanced process control and optimisation. This EPSRC project will allow Cambridge and Liverpool University to use their combined expertise and proven technology, steered by a world leading supplier of manufacture equipment (Bosch, Liverpool, UK) and a global pharmaceutical company (Pfizer, Sandwich, UK) and supported by academia (Professor De Beer, Ghent University, Belgium), a technology SME (TeraView, Cambridge, UK) and with additional insight from the regulators (Dr Wu, FDA), to provide a highly advanced manufacturing capability currently not available to the industry.