AI and data science in the age of COVID-19
Tuesday 24 Nov 2020
Time: 09:30 - 16:30
Bio
Andrew Dowsey is Chair of Population Health Data Science at the University of Bristol, a joint post between the Department of Population Health Sciences and Bristol Veterinary School, where he is Research Director for Bristol Veterinary School. His data science expertise lies at the interface of health sciences and engineering, having previously held positions in the Department of Computing and Institute of Biomedical Engineering, Imperial College London (EPSRC Postdoctoral Fellowship), the Institute of Human Development, University of Manchester (Lecturer), and the Department of Electrical Engineering and Electronics, University of Liverpool (Reader), together with visiting appointments at the University of Texas MD Anderson Cancer Centre and the Conway Institute of Biomolecular and Biomedical Research, University College Dublin.
Andrew’s research focus is on the facilitation and acceleration of health sciences research through novel statistical modelling and machine learning methodology, as well as data collection and management platforms. His team works in highly multidisciplinary environments to both support and lead investigations across three strands of research: Profiling the protein and metabolite content of biological fluids and tissues using mass spectrometry proteomics and metabolomics for understanding disease mechanism, and to discover biomarkers for clinical diagnostics; Constructing secure research platforms from population-level agricultural, environmental and health data to inform policy on One Health challenges such as global food security and the transmission of anti-microbial resistance; Intensive longitudinal health, activity and behavioural monitoring at the level of individual animals and groups of animals.
Research interests
Andrew is leading implementation of the John Oldacre Centre for Sustainability and Welfare in Dairy Production at Bristol Veterinary School, with the aim of create the world’s most intensively monitored livestock cohort at our dairy farm. The Centre will provide blanket 24/7 surveillance of each animal through video tracking, activity wearables, measurements of individual feed and emissions, environmental monitoring, and recording of veterinary assessments. While such infrastructure is key for developing future ‘precision farming’ approaches for optimising farm management in the livestock industry, our Centre is particularly focussed on harnessing this data to tackle fundamental grand challenges affecting human and animal health globally. These include the need to better understand what constitutes and predicts animal resilience, health and welfare, and what factors damage sustainability by increasing the total economic and social burden with respect to competing land use, climate change, and the effects of antimicrobial resistance.
The aim of Andrew’s work at the Turing is to build the foundations for a comprehensive open research data platform for the John Oldacre Centre. This involves several interesting data science challenges around the robust monitoring of individual animals and the integration of heterogenous data streams, with many more opening up once the platform is achieved. In this Turing project, we will work with Dr Tilo Burghardt, Department of Computer Science, University of Bristol, on extending his deep learning methodology for species localisation and individual animal biometrics robust to the movements and interactions of our 200+ Holstein cattle.