Imagine you are standing on a street corner in a city. Close your eyes: what do you hear? Perhaps some cars and busses driving on the road, footsteps of people on the pavement, beeps from a pedestrian crossing, rustling and clonks from shopping bags and boxes, and the hubbub of talking shoppers. You can do the same in a kitchen as someone is making breakfast, or as you are working in a busy office. Now, following the successful application of AI and machine learning technologies to the recognition of speech and images, we are beginning to build computer systems to tackle the challenging task of "machine listening", to build computer systems to automatically analyse and recognize everyday real-world sound scenes and events. This new technology has major potential applications in security, health & wellbeing, environmental sensing, urban living, and the creative sector. Analysis of sounds in the home offers the potential to improve comfort, security, and healthcare services to inhabitants. In environmental sound sensing, analysis of urban sounds offers the potential to monitor and improve soundscapes experienced for people in towns and cities. In the creative sector, analysis of sounds also offers the potential to make better use of archives in museums and libraries, and production processes for broadcasters, programme makers, or games designers. The international market for sound recognition technology has been forecast to be worth around £1bn by 2021, so there is significant potential for new tools in "AI for sound" to have a major benefit for the economy and society. Nevertheless, realising the potential of computational analysis of sounds presents particular challenges for machine learning technologies. For example, current research use cases are often unrealistic; modern AI methods, such as deep learning, can produce promising results, but are still poorly understood; and current datasets may have unreliable or missing labels. To tackle these and other key issues, this Fellowship will use a set of application sector use cases, spanning sound sensing in the home, in the workplace and in the outdoor environment, to drive advances in core machine learning research. Specifically, the Fellowship will focus on four main application use cases: (i) monitoring of sounds of human activity in the home for assisted living; (ii) measuring of sounds in non-domestic buildings to improve the office and workplace environment; (iii) measuring sounds in smart cities to improve the urban environment; and (iv) developing tools to use sounds to help producers and consumers of broadcast creative content. Through this Fellowship, we aim to deliver a step-change in research in this area, bringing "AI for Sound" technology out of the lab, helping to realize its potential to benefit society and the economy.

The ageing of the world population has had a devastating impact on the prevalence of people with brain disorders. The most common brain disorder with age is dementia - a neurodegenerative disease that leads to cognitive impairment that progressively affects activities of daily living erodes independence and impairs quality of life. The leading cause of dementia is Alzheimer's disease, accounting for 60-70% of all dementia cases1. There are approximately 50 million people with dementia worldwide, and this number is projected to increase to 152 million by 20502. In the UK there are approximately 850,000 people with dementia, and this number is projected to increase to 1.6 million by 2040 (translating to 1 new dementia case every 3 minutes). The global costs of dementia are estimated to be US$1 trillion annually2. The estimated cost of dementia care in the UK is £35 billion, which is projected to rise sharply to £95 billion by 2040. At every given time, about one out of four beds in the NHS hospitals is occupied by a patient with dementia3, thus impeding care for other medical conditions. During the last decades, large-scale efforts to delay or stop the progression of dementia due to Alzheimer's disease via pharmacological interventions have failed to produce viable treatment. This project will develop a technology that aims to slow or reverse the progression of Alzheimer's disease by boosting the resilience to the pathology in the most vulnerable regions at the early stages of the disease. Our approach is based on non-invasive electrical stimulation of the activity in those vulnerable structures to build up their intrinsic metabolic and energetic functionalities, in a way that is conceptionally similar to how exercise builds up the metabolic and energetic functionalities in the muscles. To non-invasively stimulate the activity at the target brain structures which are often at deep locations, we will use a novel method, called temporal interference (TI) stimulation, that we recently discovered. We have already shown that TI stimulation can be used to change the activity in the hippocampus, a deep brain structure that is critical for memory and cognitive function and strongly affected in the early stages of Alzheimer's disease, in an animal model and in healthy humans. In this project, we will address the most critical engineering challenges to develop our concept to a reliable and precise non-invasive deep brain stimulation technology that can be deployed in large-scale clinical testing. In addition, we will test and iteratively improve the effect of the temporal interference stimulation on the pathology of the hippocampus in animal models of Alzheimer's disease. Finally, we will start developing the pathway to translate the technology to a viable healthcare treatment with affordable and wearable hardware that can also be deployed at the patients' home. The temporal interference brain stimulation technology with its capability to target arbitrary deep brain structures will provide a platform for developing therapies for multiple brain disorders underpinned by aberrant activity in those structures. The development of such a disruptive technology will place the UK at the frontiers of the neurotechnology industry that is poised for the fastest growth in the medical industry. 1. Livingston, G. et al. The Lancet (2017) 2. Patterson, C. World Alzheimer Report 2018, London, UK (2018). 3. Alzheimer's Society (2009).

Sleep is an essential requirement of all animal life, including humans, who spend around a third of their life asleep. It is controlled by an internal biological clock, our circadian rhythm, which controls the timing of when we rest, sleep, and are active. We do not understand why humans need to sleep, however, it is becoming clear that sleep and circadian rhythms are linked to cognition (brain functions such as thinking and memory). Sleep and circadian rhythms are disrupted in neurological diseases where the brain is gradually damaged over time, called neurodegenerative diseases. These diseases include Alzheimer's disease, the most common cause of dementia worldwide and a leading cause of death. Alzheimer's disease causes alterations in sleep and circadian rhythms. These changes can occur before other symptoms and may help us identify those who will develop these diseases. Early diagnosis is essential as it enables treatment before damage to the brain is too widespread. Evidence is also emerging that abnormal sleep and circadian rhythms may be part of the cause of Alzheimer's disease. We do not know why abnormal sleep and circadian rhythms emerge, how they change over time, or how they relate to cognition in the long term. It is unclear if disturbed circadian rhythms are early signs of disease, or if they precede and cause neurodegenerative diseases. To answer these questions we must analyse sleep and circadian rhythms in detail. Bed mats placed under the mattress, and worn devices, called actigraphy, allow us to closely monitor sleep and circadian rhythms in patients' own homes. The genes and sleep-promoting chemicals which control sleep and circadian rhythms can be detected in the blood and spinal fluid. Advanced brain imaging, blood and spinal fluid analysis allows us to detect signs of damage and neurodegenerative diseases in the brain. The Medical Research Council National Survey for Health and Development (NSHD) offers a unique opportunity to examine the effects of abnormal sleep and circadian rhythms on cognition and neurodegenerative disease. It has followed 5362 people born in Britain on the same week in 1946, with regular questionnaires throughout their lives. The Insight 46 study has recruited is recruiting a further 872 people from NSHD to have advanced brain imaging, detailed cognitive assessments, actigraphy, genetic testing, and sampling of their blood and spinal fluid to analyse for evidence of neurodegeneration. 250 people will undergo repeat assessments, and 100 will have at least 6 months of bed mat analysis. These repeated and prolonged assessment allow particularly detailed examination of changes in sleep and circadian rhythms over time, and how they relate to progressive changes in cognition, brain imaging, and in the blood or spinal fluid. We will also examine the chemicals that control sleep and circadian rhythms in the spinal fluid of 375 participants, and examine how the genes that control sleep or make you more likely to develop Alzheimer's disease relate to sleep, circadian rhythms, and signs of neurodegenerative disease in old age. One fifth of those within Insight 46 already have early signs of Alzheimer's disease. For my PhD I plan to examine how sleep and circadian rhythms relate to cognition and neurodegenerative disease. Using sleep and circadian rhythm questionnaire data from birth to the seventh decade with sleep data from actigraphy and bed mats, I will examine sleep and circadian rhythms over the human lifetime in unparalleled detail. Combining this with the advanced brain imaging, cognitive, blood and spinal fluid data already collected from Insight 46 will allow me to explore how sleep and circadian rhythms relate to cognition and neurodegenerative disease. Finally, I will examine how this process is controlled by our genes, and sleep promoting chemicals, and whether failures in this control system are related to cognition and neurodegenerative disease.

Diseases of the brain and spinal cord are common and affect increasing numbers of people with advancing age. Conditions like Alzheimer's (AD) and Parkinson's (PD) affect a many elderly individuals, but can also present from young adult life particularly when there is a genetic cause. The same is true for other dementias and motor neuron disease (MND). These conditions are all characterised by death of specific groups of neurons (brain cells) and are referred to as neurodegenerative diseases (NDDs). They are progressive and eventually fatal and, currently, there is no cure for these disorders and a major source of disability in the population. Scientists and doctors are working to understand these and many other conditions, but the causes and methods for treating them require much more research. While valuable information can be obtained from animal models of disease, cells grown in a dish, and even computer simulations, there is no substitute for testing ideas in human tissue itself. Human tissue research is often a preliminary step before embarking on a clinical trial, especially for new treatments. For this reason, 'biobanks' have been developed that collect, characterize and store brain and spinal cord tissues from humans, both with and without nervous system diseases. These tissues, and associated data, are then provided to research projects that need to study human tissue. The goals of this proposal are several-fold. First, the current collection procedures will be greatly expanded by recruiting brain and nervous tissue donation from eight cohorts of very well studied patients with a variety of NDDs including AD and other dementias, PD, MND as well as control brains. A novel pilot study relates to Down Syndrome, a common cause of intellectual disability and early AD type dementia. We will collect brain material from embryos, fetuses, children and adults with this condition so we can understand the development of the condition in the brain over the entire lifespan. This lifespan approach can then be implemented for other NDDs particularly genetic ones. Secondly, we need to maximize the value of the donated brain material. This includes reducing the time-gap between death of the individual and acquisition of the brain, so that the tissue is as healthy as possible for research. Many types of data need to be linked to each brain sample, to provide a rich resource for the researchers, and this includes information from hospital notes, imaging data like MRIs, lab results, and microscope slides from the pathology department. Genetic data (the DNA sequence) is particularly valuable for research, and we will ensure that all samples undergo genetic analysis. The third goal of this proposal is to develop a user-friendly web-based computer platform where all samples and associated clinical information are catalogued, so they can be searched by researchers who are seeking particular tissue types, diseases or stages. Working with the MRC Dementias Platform UK informatics team we will develop such a platform. The UCL Neurodegenerative disease Human nervous tissue resource will offer a step-change in the provision of tissue for research in this area.

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.