Skip to main content
Menu

Royal Commission of 1851 Industrial Fellowships for Oxford researchers

Engineering DPhil students will use satellite data with AI to warn authorities of dam collapses; and develop computational models of type 2 diabetes to predict future health outcomes

Maral Bayaraa against a background of satellite images and Tom Waddell next to an MRI scanner

DPhil students Maral Bayaraa (left) and Tom Waddell (right)

DPhil candidates Maral Bayaraa and Tom Waddell have been awarded Industrial Fellowships from the Royal Commission for the Exhibition of 1851, which aims to encourage profitable innovation and creativity in British Industry. Approximately 10 industrial Fellowships are awarded each year to selected, exceptional graduates with the potential to make an outstanding contribution to Industry for a programme of doctoral level research.

Maral Bayaraa is working towards a DPhil in "Satellite-enabled early warning system for geotechnical structures", supervised by Dr Brian Sheil. She is developing a system that combines satellite data with artificial intelligence to warn authorities when large scale physical structures, such as dams, are on the brink of collapse. In the mining industry, ‘tailings’ dams are used to store mine waste, much of which is toxic. Maral is using tailings dams as an exemplar for her work because they are extremely prone to failure, causing the loss of human life and irreparable environmental damage, in addition to billions of dollars in losses, every year. There is an urgent need to make the dams safer through improved monitoring and better early warning systems.

From hundreds of km up in space, satellite-Interferometric Synthetic-Aperture Radar (InSAR) analysis allows detection of ground motion to a high level of precision. However, complexities in converting the raw data to actionable information has so far limited authorities’ ability to practically monitor these structures from space.

To help solve this, Maral is creating a framework for geotechnical validation of the satellite measurements by developing 3D numerical models capable of simulating the underpinning mechanics of the observed soil movements. From this, she will create deep learning experiments to create an end-to-end scalable early warning tool, tested on historical failures and other geotechnical applications. “This fellowship has given me a unique set of freedom and opportunity, where I remain embedded in industry whilst keeping hold of the freedom to publish my research open source so that it reaches as many people as possible”, she says.

 

Tom Waddell, supervised by Professor Sir Michael Brady in the Department of Oncology and Dr Ana Namburete, is developing a computational model of type 2 diabetes to help predict patient outcomes and devise personalised medical interventions. Type 2 diabetes affects 3.8 million UK adults, can lead to multiple organ dysfunction and is associated with a significantly greater risk of developing chronic conditions such as liver and cardiovascular disease.

Accurately modelling the effects of type 2 diabetes on the body’s organs is a difficult challenge as it involves a complex interplay between various biological systems. To date, no artificial intelligence-based model that incorporates MRI-derived biomarkers of multiple organs has been established to study type 2 diabetes. Tom plans to develop such a system and use it to test various hypothetical scenarios, using his model to devise personalised treatments.

Tom hopes to apply Bayesian networks – a type of graphical modelling that is well-suited to identifying causation in complex systems – and to train his model using data from magnetic resonance imaging (MRI) of internal organs. Developing a computational model of type 2 diabetes that accurately predicts future health outcomes will significantly benefit drug development and clinical trial designs around the treatment of obesity and diabetes.

Tom says of the award, “I feel extremely privileged for the opportunity to be a part of the 1851 family. This fellowship will allow me to study for my PhD and undertake valuable work in diabetes research, for which I feel very proud and extremely fortunate.”