The importance of analysing health data collected as part of clinical care and stored in electronic health records is well-established. This has led to vital research about the occurrence and progression of disease, treatment effectiveness and safety, and health service delivery. The current Covid-19 pandemic has demonstrated the public health need to efficiently use data collected at the point of care to rapidly understand patterns, risk factors and outcomes of emerging diseases. Much of this work comes from primary care electronic health records, where general practitioners (GPs) enter and use structured, coded healthcare data. The picture in hospitals, however, is very different. One in four people in the UK live with one or more long-term conditions like cardiovascular diseases, chronic respiratory diseases, type 2 diabetes, arthritis and cancer, which account for 70% of the NHS budget. Specialised opinion about management of long-term conditions (LTCs) is provided through hospital outpatient care. Data and insight from outpatient clinics, however, is almost entirely absent. There is, surprisingly, no national system for recording diagnoses in hospital outpatient clinics. Information about key clinical events is instead recorded in outpatient letters, which are primarily used to communicate with patients and GPs. The ways in which letters are written and their sensitive content mean that they are not available for larger-scale "secondary use", i.e. to support clinical practice, research or service improvement. For example, shielding for the current pandemic relied on hospital clinical teams going through patient letters manually to identify those who needed shielding based on free-text information about diagnoses and medications, with clear time constraints and risks to under- and over-shield patients. Natural language processing (NLP) and text mining develop computer algorithms to automatically extract relevant information from free-text documents. This project will establish a partnership between academia, secondary care and industry to develop a standards-based information management framework to safely unlock information stored in outpatient letters, link it with other health data and demonstrate its impact and benefits through two case studies. We will develop new methods to extract key clinical events from letters and represent their details (e.g. medication used, duration of symptoms) in a computerised form so that it can be easily accessed. In doing so, we will use the NHS-adopted standards so that the outpatient letters can be linked to other hospital databases and do not live in their own silo. The protection of sensitive data that potentially appear in outpatient data is a prime concern, so we will develop clear rules on who and how can access such data, in particular considering that third parties (e.g. industry) may need to access that data for developing their tools. These rules will be developed in a close collaboration between patient representatives, clinicians and specialists to ensure safeguards, public trust and transparency of decision making. We will demonstrate the potential impact of the proposed methods through two case studies with our clinical and business partners. Our first case study will demonstrate how the proposed models can assist in timely, efficient, dynamic and transparent identification of patients for shielding in a pandemic, or for vaccination prioritisation. In the second case study, we will illustrate how the same information can be used address important gaps in our knowledge about health and care, including, for example, disease prevalence and drug utilisation patterns. All outputs will be developed in a way that can be scaled beyond the single clinical site and single speciality.

PhD projects will be organised in three central themes that represent the core of our programme. The themes are aligned to the strategic priorities of our NHS partners and the overall vision of the CDT: A. AI-enabled diagnostics or prognostics [lead; McKendry]. Deep learning - the subset of machine learning that is based on a network structure loosely inspired by the human brain - enables networks to learn features from clinical data automatically. This gives them the ability to model complex non-linear relationships and such AI methods have found application in clinical diagnosis using either parameters typically embedded in an electronic health record (like blood test results) or the images produced during radiographic exams or in digital pathology suites. This theme will help us create, initiate and deploy academic research projects centred on clinical use cases of direct applicability in the hospitals where our Centre is based. Example projects might include the detection of radiology abnormality; characterisation of tissues and tissue abnormality (e.g. cancer staging); or the serial monitoring of disease. B. AI-enabled operations [lead; Marshall] The proximity of our Centre to the end-users of health technology prompts a second focus, on the use of AI methods to optimise care processes and pathways. We will ensure that our projects are academically focused, but will seek to create new approaches to investigate and characterise the performance of hospitals systems and processes - such as the flow of patients through emergency departments, AI-enabled projects that might shorten time-to-treatment or cancer waits. This will be the most translationally focused theme, seeking to surface and address key use cases of the greatest academic interest. C. AI-enabled therapeutics [lead; Denaxas]. Our final theme is forward looking; the use of deep learning and other AI methods in therapeutic inference or even in a therapy itself. AI methods may be most applicable here in mental health, where deployment of 'talking therapies' is as efficacious through the internet or telephony as face-to-face; or in the development of 'avatar therapies' such as that recently proposed at UCL for hallucinations. But a wide variety of research projects are conceivable, including rehabilitation following stroke; or indeed the use of AI monitoring of radiological change as a proxy endpoint for drug trials. This theme will help us focus cutting-edge work in our Centre around such use cases and novel methodology. The UK leads in the development of artificial intelligence technologies, investing around $850M between 2012-16, the third highest of any country. This has catalysed significant UK involvement of major global technology companies such as Alphabet and Apple, the creation of new UK-based AI companies such as Benevolent AI and DeepMind (both partners in our Centre) and the emergence of a vibrant UK SME community. 80% of AI companies on the UK Top 50 list are based in London, most with 30 minutes travel from UCL. Many of the most successful AI companies now focus on the application of AI in health, but the successful application of AI technologies such as deep learning has three key unmet needs; the identification of clinically relevant use cases, the availability of large quantities of high quality labelled data from NHS patients, and the availability of scientists and software engineers with the requisite algorithmic and programming skills. All three are addressed by our CDT, its novel NHS-embedded approach to training, linked to primary and social care and with close involvement of commercial partners, structured internships and leadership and entrepreneurship. This will create an entirely new cadre of individuals with both clinical knowledge and algorithmic/programming expertise, but also catalyse the creation and discovery of new large labelled datasets and exceptional clinical use cases informed by real-world clinical care.