The project develops new computer science technology for modelling the progression of a disease or developmental process. It pioneers the use of state-of-the-art generative modelling and learning techniques to address this problem. It demonstrates the new approach by addressing questions of intense current interest in neurology: what is the sequence of clinical and pathological decline in two important diseases, Alzheimer's disease (AD) and fronto-temporal dementia (FTD), and how does it vary over the population? The methodological development introduces new and general-purpose techniques in computer science and the experimental work adds fundamental new knowledge in neurology. The progression is the sequence of events that occurs as the disease or process advances. All diseases have an associated set of symptoms and pathologies. For example, AD causes loss of memory, personality changes, brain shrinkage, and deposits of abnormal proteins. However, other neurological diseases share many of these same occurrences. An additional fundamental characteristic that distinguishes diseases is the order in which the symptoms and pathologies appear. Knowledge, or a model, of this disease progression supports early diagnosis, which can maximize the effect of a treatment. It also provides insight into disease mechanisms that can accelerate development of the treatments. Furthermore, an effective model helps construct robust staging systems, which enable clinicians to tailor treatment and care plans for individual patients: so called "personalized medicine". Modelling disease progression, however, is a major challenge. First, the sequence of events can vary substantially among patients; monitoring a few individuals closely does not capture the variation over the larger population. Second, such close monitoring is often impossible, because the necessary examinations are too expensive or invasive to perform regularly. Thus, models must come from more cross-sectional data obtained from many patients each making a few irregular visits to a clinic. Very large data sets of this kind are available and contain a wealth of information, but current techniques for mining that information remain crude and do not exploit the available data effectively. The investigators on this project recently introduced a new computational approach to disease progression modelling: the event-based model. Unlike standard models, it learns the sequence of events directly from a large cross-sectional data set without requiring a-priori staging or ordering of the patients. Preliminary results using small data sets from genetically confirmed disease cohorts demonstrate the uniquely rich description of disease progression the new approach can provide. However, application to larger and less-controlled data sets, where the real interest lies, presents major new challenges. This project develops the event-based model from proof-of-concept to practical research tool. It then demonstrates the tool focussing on applications in neurological disease, although long-term applicability is much wider. In particular, we construct detailed models of the progression of AD and FTD, their variability over the population, and the influence of factors such as genetic profile. Finally, the project initiates exploration of the wider family of computational models of disease progression and their potential to extract new and fundamental information. For example, we introduce new models that potentially reveal disease subtypes, provide disease-staging systems, and highlight potential causal relationships among events. The new model-based approach has the potential to revolutionize the way we think about disease progression and thus to make a major impact in diagnosis, disease management, and treatment development for some of the most devastating and widespread medical problems facing us today. The project initiates a long-term effort towards these ends.

This programme aims to change the way medical imaging is currently used in applications where quantitative assessment of disease progression or guidance of treatment is required. Imaging technology traditionally sees the reconstructed image as the end goal, but in reality it is a stepping stone to evaluate some aspect of the state of the patient, which we term the target, e.g. the presence, location, extent and characteristics of a particular disease, function of the heart, response to treatment etc. The image is merely an intermediate visualization, for subsequent interpretation and processing either by the human expert or computer based analysis. Our objectives are to extract information which can be used to inform diagnosis and guide therapy directly from the measurements of the imaging device. We propose a new paradigm whereby the extraction of clinically-relevant information drives the entire imaging process. All medical imaging devices measure some physical attribute of the patient's body, such as the X-ray attenuation in CT, changes acoustic impedance in ultrasound, or the mobility of protons in MRI. These physical attributes may be modulated by changes in structure or metabolic function. Medical images from devices such as MR and CT scanners often take 10s of seconds to many minutes to acquire. The unborn child, the very young, the very old or very ill cannot stay still for this time and methods of addressing motion are inefficient and cannot be applied to all types of imaging. Usually triggering and gating strategies are applied, which result in a low acquisition efficiency (since most of the data is rejected) and often fail due to irregular motion. As a result the images are corrupted by significant motion artifact or blurring.Accurate computational modeling of physiology and pathological processes at different spatial scales has shown how careful measurements from imaging devices might allow the clinician or the medical scientist to infer what is happening in health, in specific diseases and during therapy. Unfortunately, making these accurate measurements is very difficult due to the movement artifacts described above. Imaging systems can provide the therapist, interventionist or surgeon with a 3D navigational map showing where therapy should be delivered and measuring how effective it is. Unfortunately image guided interventions in the moving and deforming tissues of the chest and abdomen is very difficult as the images are often corrupted by motion and as the procedure progresses the images generally diverge from the local anatomy that the interventionist or surgeon is treating.Our programme brings together three different groups of people: computer scientists who construct computer models of anatomy, physiology, pharmacological processes and the dynamics of tissue motion; imaging scientists who develop new ways to reconstruct images of the human body; and clinicians working to provide better treatment for their patients. With these three groups working together we will devise new ways to correct for motion artifact, to produce images of optimal quality that are related directly to clinically relevant measures of tissue composition, microscopic structure and metabolism. We will apply these methods to improve understanding of disease progression; guide therapies and assess response to treatment in cancer arising in the lung and liver; to ischaemic heart disease; to the clinical management of the foetus while still in the womb; and to caring for premature babies and young children.

The UKRI CDT in Artificial Intelligence (AI) for Healthcare will be the world's leading centre for PhD training of the next-generation innovators in AI applied to Healthcare. There is a unique role for AI in healthcare by providing more accurate decisions faster while reducing cost and suffering across society. AI in healthcare needs and drives current AI research avenues such as interpretable AI, privacy-preserving learning, trust in AI, data-efficient learning and safety in autonomy. These are key due to the immediate impact on life and health for users depending on AI for healthcare support. Healthcare applications require many AI specialists that can apply their skills in this heavily regulated domain. To address this need, we propose to train in total 90+ PhD students including 16 clinical PhD Fellows in five cohorts of 18+ PhDs, which will establish a new generation of cognitively diverse AI researchers with backgrounds ranging from computer science, psychology to design engineering and clinical medicine. The CDT focus areas arise from our early engagement in AI research and collaboration with clinicians, partnered technology companies and patient organisations, reflecting the healthcare areas of the UK industrial strategy. The Centre is grouped into 4 complementary healthcare themes and 4 cross-cutting AI expertise streams. The 4 healthcare themes are: (1) Productivity in Care: making healthcare provision more efficient and effective by increasing the productivity of doctors and nurses; (2) Diagnostics & Monitoring: developing AI-based diagnostics & monitoring that can detect disease earlier and monitor health with more precision; (3) Decision support systems: AI-based decision support systems that will support e.g. freeing up doctors' time to focus on the patient or can accelerate the development of novels drugs and treatments and empowering patients to be active agents within the decision-making by explaining, and (4) Biomedical discovery: driven by AI that accelerates drug discovery and linking genome, microbiome and environment data to discover novel disease mechanisms and treatment pathways. The themes are linked by 4 cross-cutting AI expertise streams: a. Perceptual AI technology enables to perceive, structure, and recognise from sensory data clinically relevant information. b. Cognitive AI technology mimics the reasoning, i.e. cognitive process, of healthcare specialists. c. Assistive AI technology supports clinicians with decision making as well as patients directly d. Underpinning AI technologies are driving factors for clinical and patient-focused AI innovations and will be enabling AI methodologies to operate beyond the currently possible. Our unique cohorts will benefit from an integrated training program and co-creation process with industry and patient organisations. PhD training is split into three phases that provide underpinning skill training (Foundation phase), research training (Research Phase) and finally drive PhD impact (Impact phase). During the Impact phase, the students will either (1) commercialising their research through a mentored start-up route (incubator partners), (2) deploying their technology in a clinical trial (two NIHR biomedical research centre (BRC) partners), or (3) testing their work in person through an NHS honorary contract (three NHS trusts as partners). Bespoke training will be created, such as AI bias & ethics, security, trust, inclusivity, differential privacy, transparency, accessibility and usability, service design, global inclusivity, healthcare treatments, clinical statistics and data regulation, Healthcare technology regulation, and technology commercialisation. We offer an exit Strategy (month 9-12) through a master's degree. The centre will place special emphasis on research that explores diversity in AI for healthcare research, including services to underserved communities and minority-specific care requirements.

Medicine is undergoing a simultaneous shift at the levels of the individual and the population: we have unprecedented tools for precision monitoring and intervention in individual health and we also have high-resolution behavioural and social data. Precision medicine seeks to deploy therapies that are sensitive to the particular genetic, lifestyle and environmental circumstances of each patient: understanding how best to use these numerous features about each patient is a profound mathematical challenge. We propose to build upon the mathematical, computational and biomedical strengths at Imperial to create a Centre for the Mathematics of Precision Healthcare revolving around the theme of multiscale networks for data-rich precision healthcare and public health. Our Centre proposes to use mathematics to unify individual-level precision medicine with public health by placing high-dimensional individual data and refined interventions in their social network context. Individual health cannot be separated from its behavioural and social context; for instance, highly targeted interventions against a cancer can be undermined by metabolic diseases caused by a dietary behaviour which co-varies with social network structure. Whether we want to tackle chronic disease or the diseases of later life, we must simultaneously consider the joint substrates of the individual together with their social network for transmission of behaviour and disease. We propose to tackle the associated mathematical challenges under the proposed Centre bringing to bear particular strengths of Imperial's mathematical research in networks and dynamics, stochastic processes and analysis, control and optimisation, inference and data representation, to the formulation and analysis of mathematical questions at the interface of individual-level personalised medicine and public health, and specifically to the data-rich characterisation of disease progression and transmission, controlled intervention and healthcare provision, placing precision interventions in their wider context. The programme will be initiated and sustained on core research projects and will expand its portfolio of themes and researchers through open calls for co-funded projects and pump-priming initiatives. Our initial set of projects will engage healthcare and clinical resources at Imperial including: (i) patient journeys for disease states in cancer and their successive hospital admissions; multi-omics data and imaging characterisations of (ii) cardiomyopathies and (iii) dementia and co-morbidities; (iv) national population dynamics for epidemiological and epidemics simulation data from Public Health; social networks and (v) health beliefs and (vi) health policy debate. The initial core projects will build upon embedded computational capabilities and data expertise, and will thus concentrate on the development of mathematical methodologies including: sparse state-space methods for the characterisation of disease progression in high-dimensional data using transition graphs in discrete spaces; time-varying networks and control for epidemics data; geometrical similarity graphs to link imaging and omics data for disease progression; stochastic processes and community detection from NHS patient data wedding behavioural and social network data with personal health indicators; statistical learning for the analysis of stratified medicine. The mathematical techniques used to address these requirements will need to combine, among others, ingredients from dynamical and stochastic systems with graph-theoretical notions, sparse statistical learning, inference and optimisation. The Centre will be led by Mathematics but researchers in the Centre span mathematical, biomedical, clinical and computational expertise.

The term "dementia" is used to describe a syndrome that results, initially, in cognitive function impairment and in many cases, a descending staircase of psychological dysfunction, leading eventually to death. It is a major socio-economic challenge with care costs approaching 1% of global GDP. Several conditions that lead to serious loss of cognitive ability are grouped under this syndrome, including Alzheimer's disease (AD), Vascular Dementia (VaD), Frontotemporal Dementia, etc. A high publicity announcement was made in 2012, by the Prime Minister, emphasising the high priority that should be given to dementia-related research and that funding will more than double in the immediate future, to partially remedy the fact that the overwhelming impact of the syndrome has been over-looked (Guardian, 26/3/12). On Dec 2013, the G8 Summit hosted in London brought together G8 ministers, researchers, pharmaceutical companies, and charities to develop co-ordinated global action on dementia. Dementia has marked adverse effects on the quality of life of tens of millions of people (both patients and carers) and exerts tremendous pressure on healthcare systems, especially when clear trends towards an ageing population, changing environmental influences and contemporary lifestyle choices are considered. Ca. 35M people suffer from dementia worldwide, a figure to quadruple by 2050. Europe and North America share a disproportionally high burden: the effects of ageing are particularly stark for these regions, exacerbating the healthcare provision implications. The Clinical Relevance: Vascular Cognitive Impairment (VCI). VCI defines alterations in cognition attributable to cerebrovascular causes, ranging from subtle or fixed deficits to full-blown dementia. VCI is a wide and accepted term referring to the "syndrome with evidence of clinical stroke or subclinical vascular brain injury and cognitive impairment affecting at least one cognitive domain", with resulting VaD being its most severe form. VaD is responsible for at least 20% of dementias, second only to AD, with a prevalence doubling every 5. 3 years. Several trials examined cholinesterase inhibitors for the treatment of vascular dementia, but the benefits are very modest, except in the individuals with a combination of AD and VaD. Vascular changes result in white matter (WM) damage (leukoaraiosis), which profoundly affect the fidelity of the information transfer underlying brain function and cognitive health8. Cerebral Magnetic Resonance Imaging (MRI) of Diffusion and Perfusion. MRI is a medical imaging technique affording non-invasive investigation of anatomy and tissue function, which is particularly suited to studying cognitive disorders due to its sensitivity and reliability. Our main interest is to characterise vascular and non-vascular tissues using quantitative diffusion and perfusion MR. Our overall aim is to characterise and quantify early differential alterations in brain blood transport and subsequent microstructural tissue damage using one-stop-shop perfusion/diffusion MR GSI incorporating novel MR signal models and optimal MR sequence design based on new human brain histomorphometric data in health and disease.