Researchers investigating quantum chaos, medical imaging, new materials and stubborn infections have received ERC Starting Grants.
Starting Grants recognise talented early-career scientists who show potential to be research leaders, awarding up to €1.5 million over five years.
Quantum chaos and technology
This is a game changer for my research, allowing me to significantly expand my team and get ahead in this exciting and fast developing new area.
– Dr Eva-Maria Graefe
Dr Eva-Maria Graefe, from the Department of Mathematics, is leading the Universality and Chaos in PT-symmetric quantum systems (CHAOS-PIQUANT) project. This area of quantum physics has been gaining a lot of attention, but there is still much to be explored in its theory, experimentation and application.
Quantum physics governs the behaviour of very small particles, which differs from that of our large-scale everyday world. However, microscopic particles make up the stuff of our everyday world, so why physics at the two scales should act so differently opens up a lot of questions.
In particular, Dr Graefe investigates differences in how certain systems at the two scales react to chaos – large-scale changes caused by small initial fluctuations. A classic illustrative example of chaos is a butterfly flapping its wings in Brazil causing a tornado in Texas.
The research project focuses on the fingerprints and possible applications of chaos in open quantum systems with balanced gain and loss.
At the time of her grant interview, Dr Graefe had recently returned from maternity leave, having had her second daughter just six months before. She said: “I am absolutely delighted to have been awarded an ERC starting grant. This is a game changer for my research, allowing me to significantly expand my team and get ahead in this exciting and fast developing new area.”
Fighting stubborn bacteria
Dr Sophie Helaine, based at the Centre for Molecular Bacteriology and Infection, is leading a project to investigate stubborn bacterial cells called persisters. These cells, which can remain inactive for periods before they reawaken to cause infection, are thought to be a key reason why many bacterial infections are so difficult to treat, and are resistant to many drugs as their growth is ‘paused’.
Dr Helaine said: “We will focus our efforts in understanding how these bacteria eventually wake up to lead to relapse of infections. This way, we hope to be able to design new strategies to target them and tackle bacterial persistence. Our model organism is Salmonella.
“Receiving this grant is a tremendous chance to advance our research at high pace and an amazing sign of recognition that we are on the right track. I am thrilled!”
Accelerating materials discovery
Dr Kim Jelfs, from the Department of Chemistry, will lead the Computational Molecular Materials Discovery (CoMMaD) project with her Starting Grant. CoMMaD aims to accelerate the discovery of new molecular materials, with a focus on materials that could perform separations or catalysis, and be used in sensing or photovoltaics (solar cells).
To do this, Dr Jelfs hopes to create software to screen libraries of molecular building blocks through a combination of artificial intelligence techniques including evolutionary algorithms and machine learning. The software will be designed to choose the most promising sets of building blocks and predict their molecular structures and properties, ensuring that it will be feasible to synthesize the best new materials.
Dr Jelfs was on maternity leave when she attended her interview, and like Dr Graefe, had to juggle a young child, in this case an eight-week-old baby. On gaining her grant, Dr Jelfs said: “I am thrilled to be awarded this ERC Starting Grant, and I can’t wait to get started on expanding my group’s research effort into discovering new materials.”
Medical scans assessed by artificial intelligence
The ERC funding will enable me to take a comprehensive approach to tackle major challenges in medical image analysis.
– Dr Ben Glocker
AI technology that is capable of analysing and interpreting medical scans with super-human performance will be the focus of ERC funded research carried out by Dr Ben Glocker, from the Department of Computing.
The sheer complexity and volume of data being generated by medical imaging technology and the ability to interpret and extract clinically useful information is pushing the ability of medical researchers and practitioners to the limit. This may increase the risk that signs of disease are missed.
Dr Glocker and his team aim to develop AI technology based on machine learning that can sift through the mountain of data to provide detailed insights into complex diseases, which may not easily be recognised by human experts.
The AI technology will have the ability to analyse large datasets on populations and carry out multiple analyses of scans simultaneously, such as the brain and the heart. It will then cross reference the results with other important factors implicated with disease such as age, lifestyle influences and genetics to provide detailed insights.
Dr Glocker said: “The ERC funding will enable me to take a comprehensive approach to tackle major challenges in medical image analysis. My team will aim to build powerful and trustworthy algorithms to automatically detect subtle signs of disease.
“Current methods behave more like black boxes, but we need to better understand how those methods work, when they work and when they get it wrong. This is important to build the trust among doctors, patients and policy makers for using AI in healthcare, which has the potential to improve diagnosis, therapy and treatment of complex diseases.”
Article text (excluding photos or graphics) available under an Attribution-NonCommercial-ShareAlike Creative Commons license.
Photos and graphics subject to third party copyright used with permission or © Imperial College London.
Communications and Public Affairs
Communications and Public Affairs
Leave a comment
Your comment may be published, displaying your name as you provide it, unless you request otherwise. Your contact details will never be published.