There is an urgent, unmet need for reliable, intelligent systems that can monitor patient condition in the home, and which can help patients manage long-term conditions. Delays in recognition of the changes in physiological state worsen outcomes and increase healthcare costs. The ASPIRE programme uses chronic obstructive pulmonary disorder (COPD) as an exemplar, which affects over 210 million people globally. This condition costs the National Health Service over £800 million each year, over half of which is spent treating patients in hospital, rather than caring for them in their homes. Intelligent monitoring systems are required to address the needs of patients with long-term conditions in their homes. However, no wearable systems have penetrated into clinical practice at scale, due to: (i) poor tolerance of existing wearable devices for monitoring; (ii) a lack of robustness in the estimates of the vital signs that wearable sensors produce; (iii) very limited battery life that requires batteries to be re-charged at a rate that prevents their use on a large scale; and (iv) limited subsequent use of the data for helping the patient understand and manage their condition. We propose to develop an "intelligent" home-based system, with smart algorithms embedded within lightweight healthcare sensors, to overcome these limitations. Our novel work will incorporate next-generation machine learning algorithms to combine information from healthcare sensors with information from GP and hospital visits. This will enable the system to learn "normal" health condition for individual patients, with knowledge of other conditions from which they may be suffering, and which can then make recommendations to the patient concerning self-management of their condition. This work will include close working with world-leading clinicians to ensure that the recommendations provided by the system are correct for the individual patient.

Cardiovascular diseases (CVDs), such as strokes and heart attacks, are the leading causes of death amongst women globally. A particular group of women at high risk of cardiovascular disease are those who experience a pregnancy complicated by gestational diabetes mellitus (GDM) and/or a hypertensive disorder of pregnancy (HDP). Following a pregnancy affected by GDM, within 5 years up to 50% of women will develop T2DM; and following HDP, 30% of women will develop hypertension. Both these conditions greatly increase risk for CVD, however with timely detection and management these risks can be greatly reduced. The importance of breaking this link between high-risk pregnancy and CVD is widely acknowledged, yet to date there have been no trials demonstrating this can be achieved, and importantly whether it can be done affordably and at scale. Three key actions are needed: (i) effective primary prevention; (ii) regular screening, and (iii) evidence-based management when disease is detected. India is experiencing an epidemic of type 2 diabetes mellitus (T2DM) and hypertension. 73 million people have diabetes, and 207 million hypertension (2017 data). Rates of GDM and HDP are high, affecting 20% and 10% of pregnancies, respectively. There is an urgent need for effective and affordable preventative strategies to reduce the economic, social and health consequences of these conditions for women in India. In the UK, GDM and HDP are the commonest complications of pregnancy. After a pregnancy with GDM, women should undergo screening with their GP 6 week after birth for persistent high blood glucose. Attendance however is generally poor, with rates between 30-70% across the country. Following HDP, evidence is needed to guide care. The Fellowship will enable me to lead connected programs of work across two countries: India and the UK, determining the role digital innovations could play to deliver post-partum interventions in women globally. I will conduct two clinical studies, with active engagement with policy makers, clinicians, digital health companies and social enterprises throughout the Fellowship. SMART Health is a digital platform, developed by the George Institute for Global Health, that has been implemented in India, Indonesia, China, and Myanmar, to improve detection and management of diabetes and hypertension. The platform is aimed at rural community health workers and primary care doctors, enabling task shifting, clinical decision support, automated referral, SMS reminders, and patient tracking. Since 2017, I have been leading the group adapting this platform to improve the detection and management of anaemia, GDM and HDP in pregnant women living in rural India: SMART Health Pregnancy (SHP). Through this Fellowship I will extend SHP to facilitate prevention, screening, and early treatment of hypertension and T2DM in the years after a pregnancy complicated by GDM and/or HDP. The effect of this on achieving target blood pressure and blood glucose control after high risk pregnancy will be assessed in a large clinical trial in rural India, following 960 women for 5 years. In the UK, I was part of the team of clinicians and researchers in Oxford who developed a remote monitoring system for women with GDM (GDmHealth). We demonstrated in a clinical trial that this approach was safe, convenient and preferred by women and health workers. The technology was licensed to a commercial company (Sensyne) in 2018, and since then thousands of women have benefited from this innovation across the UK. Through this Fellowship I will lead a program of work adapting this approach for women in the year after birth, assessing whether remote monitoring improves screening attendance, deliver effective lifestyle support, offer a potential cost savings to the NHS, whilst being acceptable and more convenient for new mums. Theis approach could improve their health, for future pregnancies and lifelon

Over 20 million people in the UK live with rheumatic and musculoskeletal diseases (RMD), and inflammatory arthritis (IA) is a major subdivision of RMD causing joint inflammation leading to damage. IA causes long-term pain, disability and incurs substantial personal and societal costs. There is also an estimated 59% increase in diagnosed IA cases between 2004 and 2020 in the UK which has important implications for health services. Rheumatology departments accounted for approximately 9% of the average NHS trusts total medication spend in 2019/2020. There are still significant unmet needs in the IA patient pathway, especially in IA detection and flare management. IA presents with non-specific symptoms and there is currently no diagnostically definitive single biomarker for IA. Early detection is critical but challenging, and delay in detection and late referral often result in loss of the window of opportunity when effective treatment should start and delays can lead to disability and associated unemployment. For patients who are diagnosed with IA, IA outcomes and activities such as flare-up are very heterogeneous in their manifestations between individual patients. Real-world data from The National Early Inflammatory Arthritis Audit showed inequality in care for rheumatology patients from minority ethnic groups. A lower proportion of ethnic minority patients achieved disease remission compared to white patients. UN4 Finally, weather is another contributing factor of IA flare heterogeneity. Despite significant unmet needs, RMD, especially IA, is still an underexplored area of real-world ML application in comparison with other diseases. Existing ML studies do not fit for purpose of early detection in practice as they are not trained based on the data available at the point of early detection. Furthermore, although there are studies showing potential determinants of IA, there is no research, or any machine learning methods that can identify the undetected determinants-combination that can offer a useful level of prediction of IA. This is because current ML approaches still cannot handle the underlying relationships among heterogenous datasets with different data types, modalities, contexts, cohorts and levels of incompleteness. On the other hand, existing ML methods in IA, and healthcare in general, still rely on a "one-size-fits-all" paradigm rendering generic learning algorithms, suboptimal on the individual level especially as IA is known to be heterogenous in nature from the time of diagnosis. Although there are methods for explainable ML local, there is limited research to quantify and explain model prediction uncertainty and its usability in practice. For a physician to use and trust ML predictions it is critical to understand the uncertainty associated with these predictions for the individual patient. Although successful translation requires bringing together expertise and stakeholders from many disciplines, the development of ML solutions is currently occurring in silos, and there is a lack of holistic and scalable ML development pipeline. Despite all the limitations of current ML, there are huge opportunities to advance ML, especially in rheumatology applications, because rheumatology has already been leading the way in the use of virtual clinics and remote monitoring in the UK. It is now time to advance ML using data generated for real early detection and personalised management of IA. Our vision: The proposed project will develop useful and responsible machine learning methods to achieve real-world early detection and personalised disease outcome prediction of inflammatory arthritis. We will develop a holistic and scalable approach through an interdisciplinary team addressing the pressing healthcare challenges of inflammatory arthritis and the limitations of machine learning to accelerate real-world ML application in healthcare.

Every year scientific journals publish tens of thousands of articles describing findings from health research studies. However, readers and users of these articles, who include scientists, clinicians, systematic reviewers, and increasingly also patients, find many of these articles very difficult or impossible to use: many articles do not present enough information, present only selected information, or present information in a very unclear and misleading way. All this makes many papers unusable. The effort and money devoted to the research described in such an unsatisfactory manner is wasted. A simple solution to improve the completeness, accuracy and clarity of research papers is to follow reporting guidelines. Many guidelines exist that provide step by step guidance of what should be addressed in a paper reporting on a particular type of health research. These guidelines have been developed from the users' perspective and guide authors to provide minimum information a user needs to assess how well was the study done, to decide if the findings are relevant to his/her own work, and if needed to reproduce the study (ie. what was actually done and to whom, what was assessed and how, how were these findings analysed, and what they actually mean in the context of other similar studies). Although many good guidelines exist they are still not widely known and used by health scientists. Recent reviews of publications consistently show that essential information is missing from a large proportion of research articles. In this time of massive information overload it is important to have a single good quality resource where you can easily find all relevant information you need. In 2008, we launched the EQUATOR programme that aims to enhance the quality and transparency of health research. One of the most important outputs of this programme is a free online Library for Health Research Reporting that brings together all published reporting guidelines and other helpful tools that aid the writing and publication of research reports and thus improve the information provided to readers. The EQUATOR team educates scientist and journal editors, who play a key role in safeguarding the quality of published papers, to increase their knowledge of what should be included in research papers and how best to achieve it. EQUATOR also helps scientists to develop high quality reporting guidelines and conducts research investigating problems in research reporting. Our proposal outlines specific deliverables and activities for the next three years that will further advance the programme. The main outputs include: improved structure and content of our Library; development of unique EQUATOR 'signature' courses supporting rigorous research reporting; compilation of a manual for the development of robust reporting guidelines; a research report summarising the use of reporting guidelines by selected priority journals; and a database of evaluations of reporting quality of scientific papers across health research specialties. Medical journals publish large numbers of research reports that are of limited value because of crucial omissions. This waste is avoidable. The EQUATOR website and training can be compared to a well stocked and well promoted supermarket where you can get everything you need to write and publish first class research papers. The knowledge of what needs to be included in research papers that are clear and easy to use also improves the design of future research studies. Our work helps to improve usability and usefulness of published medical research and helps scientists to become outstanding research communicators.

Medical imaging techniques such as MRI have revolutionised clinical diagnosis, treatment and monitoring of disease. However, they are expensive and not readily accessible outside specialist units. Imagine if instead, there was available a high-street eye test that provided diagnostic information for a range of diseases. These diseases could be neurodegenerative diseases such as Alzheimer's, systemic diseases (diseases with wide-spread effect on the body) such as heart disease, or psychiatric conditions, such as depression. The test would be sensitive, picking-up signatures of disease before any symptoms were apparent and before irreparable damage had occurred, and allowing fine scale monitoring of changes in response to treatment. It would offer specificity, differentiating between diseases with different aetiologies but similar retinal manifestations. This would allow mechanistic understanding of disease progression, paving the way for future therapies. The key to realising this vision is the application of recent technological advances from microscopy, image and signal processing to high-resolution optical imaging of the living human retina. The retina, which is the tissue at the back of our eye, is in fact a part of the central nervous system and has long been recognised as a window to the brain and vasculature. In fact, psychiatric, neurodegenerative, and systemic diseases have been shown to have detectable correlates in the eye. However, current clinical technology cannot image individual cells, and so these diseases manifest in gross anatomical changes that cannot be distinguished amongst diseases. We will develop a non-invasive optical instrument, capable of imaging individual cells and testing their function, for sensitive and specific detection of these diseases. The technology would revolutionise point-of-care medicine by providing rapid, non-invasive diagnostics on a range of conditions, replacing costly, time-consuming current gold standard methods. Our team is a collaboration between technology developers and ophthalmic specialists, spanning engineering and medical science within partner institutions. We already have experience in human participant testing across the life-span with bespoke optical instrumentation, and extensive experience in commercialisation of technology, industrial partnership and spin-outs. The required technological components - for example, optical interferometry, adaptive optics, spectroscopic and polarisation techniques, holography, and dedicated image and signal processing - are available in the related fields of microscopy and ophthalmoscopy, but delivering an integrated instrumentation package remains a significant engineering challenge. The development phase will be vital for establishing proof-of-principle demonstrations to engage stakeholders, and to target efforts to those areas that are most likely to have 'disruptive' impact in healthcare. Stakeholders - clinicians, industry partners and patient groups - will be engaged through local NHS Trusts and teaching hospitals, existing industry networks and charities representing specific patient cohorts. During the development phase we will widen and deepen these networks. With a long-term view, we will engage at all levels of medical training - from the pre-clinical undergraduate to the established consultant. Three significant challenges facing society are the high incidence of mental health issues across the population, cardiovascular disease, and neurodegenerative diseases which disproportionately affect the elderly and are of great concern in an ageing society. Dementia and heart disease are the leading causes of death in the UK, and indeed world-wide. Faster and more effective diagnosis and treatment of such debilitating conditions will significantly improve outcomes for these patients. Widespread uptake of the technology will lead to new business growth through commercialisation.