The University of Cambridge-led study shows that children from the very poorest families, in what are already some of the lowest-income countries in the world, consistently perform worse in basic literacy and numeracy tests than those from more affluent backgrounds.

The overwhelming reason, the study found, is that poorer children are disproportionately clustered in the lowest-quality schools, which often lack even basic resources – such as textbooks, electricity, or toilets.

The researchers say that there is an urgent need to ‘raise the floor’ in global education, by focusing both national-level efforts and international aid on students from the most disadvantaged communities.

Institutions like the United Nations Educational, Scientific and Cultural Organisation (UNESCO) and the World Bank have long referred to a ‘learning crisis’ in the Global South. While growing numbers of children in low-income countries now attend school compared with previous generations, many still lack basic literacy or numeracy skills.

Until now, most analyses have looked at the factors that explain low learning outcomes in general, rather than differentiating between groups of children. But the new study suggests that there is a huge gulf between the quality of education that children from the poorest families receive compared with wealthier children, and that this is directly linked to their ability to read, write, add, or subtract, by Grade 6.

Dr Rob Gruijters, from the Research for Equitable Access and Learning (REAL) Centre, at the Faculty of Education, University of Cambridge, who led the research, said: “There is a high level of social segregation in many of these countries’ education systems. The pattern is similar to the UK, where rich children tend to go to better-resourced schools. But the differences in school quality are much more pronounced, and they are strongly linked to family background”

“Global reporting on the learning crisis often pays little attention to these inequalities, focusing instead on average differences between countries. But if we really want to fix things, there needs to be a commitment not only to investing in education, but to raising the floor: to ensuring that every school has a minimum level of support, in staffing, training, and resources.”

The study analysed data from the Programme for the Analysis of Education Systems (PASEC), a survey managed by the association of education ministries in francophone Africa. The survey assessed more than 30,000 Grade 6 students in more than 1,800 schools in 10 countries: Benin, Burundi, Burkina Faso, Cameroon, Chad, Congo (Brazzaville), Ivory Coast, Niger, Senegal and Togo. All 10 have ‘received scant attention’ in previous analyses of the learning crisis, the study says.

The data provides the pupils’ scores in basic maths and reading tests. The researchers cross-referred this with additional information about their socio-economic backgrounds, their health, and the quality of their schools; dividing each country’s sample group into fifths based on their families’ relative wealth.

Overall, pupils from the poorest 20% of families consistently performed worst in the tests, while those children who – although often poor by international standards – fell into the wealthiest 20%, consistently had the highest test scores.

Poorer students also tended to fail to reach PASEC’s Grade 6 ‘proficiency threshold’, meaning that by the time they leave primary school, many still struggle with basic sums and reading.

The researchers then explored possible reasons why this link between household wealth and performance exists. They found that differences in the quality of schooling explained almost the entire learning gap between poor and wealthier children.

Children from disadvantaged backgrounds were consistently found to be clustered in educational settings that scored low for school quality in the dataset – meaning that teachers’ own education levels were often poor, classrooms overcrowded, and critical resources and facilities, from textbooks to running water, often unavailable. Wealthier children, on the other hand, were much more likely to attend better-resourced private schools.

Importantly, in cases where children from the wealthiest 20% and poorest 20% of families attended the same school, there was almost no difference in their test results.

“The problem is that most of them are not attending the same schools, and that’s why we are seeing these learning gaps" said Dr Julia Behrman of Northwestern University, who co-authored the study. “Wealthier children learn more largely because they are going to better schools, with better resources.”

The researchers say that their assessment of the impact of socioeconomic status on learning outcomes is almost certainly conservative, as the PASEC data only covers children who reach Grade 6. In countries like Burkina Faso, Niger and Chad, where fewer than half of all children finish primary school and many never attend, the poorest children face a ‘double hurdle’: first, getting to school; and second, finding a school that is sufficiently equipped to give them a basic education.

The study therefore argues that policy initiatives and aid efforts aimed at solving the global learning crisis should focus on equalising access to learning opportunities for all children.

“One silver lining is that our research emphasises there is nothing inherent in being poor that stops children from learning,” Gruijters added. “Give them a better place to learn, with better resources, and they can do just as well as children from the wealthiest end of the scale.”

Professor Fabian is one of seven scientists from five countries honoured for breakthrough discoveries in astrophysics, nanoscience and neuroscience. 

The Norwegian Academy of Science and Letters today announced the 2020 Kavli Prize Laureates in the fields of astrophysics, nanoscience and neuroscience. This year’s Kavli Prize honours scientists whose research has transformed our understanding of the very big, the very small and the very complex. The laureates in each field will share $1million USD.

“The 2020 Kavli Prize Laureates represent truly pioneering science, the kind of science which will benefit humanity in a profound way, inspiring both current and future generations,” says Hans Petter Graver, president of The Norwegian Academy of Science and Letters.

The Kavli Prize in Astrophysics is awarded to astronomer and astrophysicist Andrew Fabian for his pioneering research and persistence in pursuing the mystery of how black holes influence their surrounding galaxies on both large and small scales. For decades, researchers have pondered the mechanics and physical processes of galaxies, and many have made discoveries that point to aspects of their inner workings; yet none has the unique vantage point of Fabian: to take a multi-scale understanding and systematically know where to look to put the pieces of the puzzle together and create the bigger picture of this vast ecosystem.

In the current cosmological paradigm, the universe is a ‘living’ system, in which the flows of gas into galaxies and black holes at their centres, and the subsequent release of energy back into the galaxies and their surroundings, all play vital roles. As the darkest objects in the universe, black holes are observed as their gravity attracts surrounding gas, dust and stars, which swirl into them at high velocities, creating intense radiation, much of it X-rays. Observational X-ray astronomy opened up access to view these and other extremely hot and energetic components of the universe, providing stunning evidence for these processes at work, unveiling how the major constituents of the universe can profoundly influence its overall evolution.

Professor Fabian employs X-ray astronomy to explore the physics of the universe. His body of work – from understanding large-scale galactic evolution to the physics of black holes at the centres of galaxies – enabled him to make connections between local conditions around supermassive black holes and the larger gas flows within and between galaxies. This research provided evidence that supermassive black holes at the heart of galaxies are the engines that drive the flow of hot gas out of the galaxy, redistributing energy through the universe and providing the building blocks for future galaxy formation.

“Andrew Fabian is one of the most prolific and influential astronomers of our time,” said Viggo Hansteen, chair of the Kavli Prize Committee in Astrophysics. “His research, breadth of knowledge and insights into the universe provided the essential physical understanding of how disparate phenomena in this ecosystem are interconnected.”

The Kavli Prize is a partnership between The Norwegian Academy of Science and Letters, the Norwegian Ministry of Education and Research and The Kavli Foundation (US). The Kavli Prize honours scientists for breakthroughs in astrophysics, nanoscience and neuroscience that transform our understanding of the very big, the very small and the very complex. Three million-dollar prizes are awarded every other year in each of the three fields. The Norwegian Academy of Science and Letters selects the laureates based on recommendations from three prize committees whose members are nominated by The Chinese Academy of Sciences, The French Academy of Sciences, The Max Planck Society of Germany, The U.S. National Academy of Sciences and The UK’s Royal Society. First awarded in 2008, The Kavli Prize has honoured 54 scientists from 13 countries – Austria, Czech Republic, France, Germany, Japan, Lithuania, The Netherlands, Norway, Russia, Sweden, Switzerland, the United Kingdom and the United States.

This year’s Kavli Prize Laureates also include Ondrej L Krivanek, who obtained his PhD from Trinity College, Cambridge, and is now based in the United States. He was awarded the Kavli Prize in Nanoscience, along with Harald Rose (Germany), Maximilian Haider (Austria), Knut Urban (Germany). 

David Julius (US) and Ardem Patapoutian (US) were awarded the Kavli Prize in Neuroscience. 

The Kavli Prize Laureates are typically celebrated in Oslo, Norway, in a ceremony presided over by His Majesty King Harald followed by a banquet at the Oslo City Hall, the venue of the Nobel Peace Prize. Due to the COVID-19 pandemic, this year’s award ceremony is postponed and will be held together with the 2022 award ceremony in September 2022.

Before COVID-19, I split my time between my Vet School office and endless meetings in the central University and in London, including at Defra and the Royal College of Veterinary Surgeons. Now I’m based in my conservatory at home, which oscillates between freezing cold and too hot! I seem to have even more meetings now, by Webex, Zoom, Skype and Teams. Talking to people on the telephone is a great release from the two dimensional world many of us now live in. 

I’m organising the Vet School’s research and policy responses to the epidemic, and working with colleagues in Cambridge Infectious Diseases, one of the University’s Interdisciplinary Research Centres, to do the same. I’m also supporting a total revision of the veterinary course examination and assessment - whilst trying to continue with my own multidisciplinary research in infection dynamics and disease control. And I’m providing a weekly hour-long ‘phone-an-expert’ service on the Jeremy Sallis Show on BBC Radio Cambridgeshire. 

I study zoonotic diseases, which are infections that spread from humans to animals. My research is mostly based in sub-Saharan Africa and India. In Ethiopia and India I’ve been working on bovine tuberculosis, and in Ghana I’m studying viruses that come from bats - like COVID-19 probably did - in order to reduce the chances of spread to human populations.  

There is an almost total lack of necessary health infrastructure in low and middle-income countries. This may result in massive mortality from COVID-19 in these places, and it is certainly likely to further emphasise health and wealth inequalities. I think this is a really major challenge in addressing the pandemic globally.

I suspect that the pandemic will further raise interest in zoonotic infections and help us to do more about them. We need far larger structural programmes to address the global challenges from these diseases. This has been a colossally neglected area. I hope that future epidemics like this can be averted through better preparation and policies based on scientific evidence. We’ve been saying this for the 15 years or so that we’ve been working on zoonotic bat viruses. Hopefully more people will listen now. 

The Cambridge research community has stepped up to this challenge in so many ways. Vaccination work has been a focus of studies within the Veterinary School. Amazing amounts of PPE were provided to Addenbrooke’s Hospital from Cambridge’s science departments. A new rapid COVID-19 test came from a Cambridge spinout. Colleagues in the Department of Engineering have been working to improve access to ventilators, and infection researchers have been supporting laboratory setups across sub-Saharan Africa with support from the Cambridge-Africa programme. 

There’s a huge ongoing contribution from Cambridge’s infectious diseases community. I’m Co-Chair of Cambridge Infectious Diseases, which has been supporting researchers with new COVID-19 research projects and establishing joint interdisciplinary seminar programmes across the University. We’ve also helped the Centre for Science and Policy (CSaP) develop its ‘Cambridge on call’ programme, which connects selected ‘Policy Fellows’ to Cambridge experts for support in developing policy responses to the crisis. And we’re working with partners to produce guidelines, for many different bodies, to reduce the risk of pandemic spread through trade in live animals. 

When this is over I’m looking forward to a holiday away from home with the family, and seeing my isolated mother. It would also be great to reinstate plans to bike across the Alps from Annecy to Nice with a friend, although I’m not quite sure how that’s all going to happen any time soon! 

James Wood is Alborada Professor of Equine and Farm Animal Science, and Head of Department of Veterinary Medicine. Listen to his overview of infectious disease modelling on Cambridge's Centre for Science and Policy podcast: Science, Policy & Pandemics Episode One.

How you can support Cambridge’s COVID-19 research

The study, led by the Scott Polar Research Institute at the University of Cambridge, used patterns of delicate wave-like ridges on the Antarctic seafloor to calculate how quickly the ice retreated roughly 12,000 years ago during regional deglaciation.

The ridges were produced where the ice sheet began to float, and were caused by the ice squeezing the sediment on the seafloor as it moved up and down with the movement of the tides. The images of these landforms are at unprecedented sub-metre resolution and were acquired from an autonomous underwater vehicle (AUV) operating about 60 metres above the seabed. The results are reported in the journal Science.

While modern satellites are able to gather detailed information about the retreat and thinning rates of the ice around Antarctica, the data only goes back a few decades. Calculating the maximum speed at which an ice sheet can retreat, using sets of these seafloor ridges, reveals historic retreat rates that are almost ten times faster than the maximum observed rates of retreat today.

“By examining the past footprint of the ice sheet and looking at sets of ridges on the seafloor, we were able to obtain new evidence on maximum past ice retreat rates, which are very much faster than those observed in even the most sensitive parts of Antarctica today,” said lead author Professor Julian Dowdeswell, Director of the Scott Polar Research Institute.

The study was carried out as part of the Weddell Sea Expedition, which set out in early 2019 to undertake a science programme and to find Sir Ernest Shackleton’s doomed ship Endurance. Although sea ice conditions at the time prevented the team from acquiring imagery of the legendary wreck, they were able to continue with their scientific work, including mapping of the seafloor close to the Larsen Ice Shelf, east of the Antarctic Peninsula.

Using drones, satellites and AUVs, the researchers were able to study ice conditions in the Weddell Sea in unprecedented detail.

Their goals were to investigate the present and past form and flow of the ice shelves, the massive floating sections of ice that skirt about 75% of the Antarctic coastline, where they act as a buttress against ice flow from inland.

Like much of the rest of the ice in the polar regions, these buttresses are weakening in some parts of Antarctica, as witnessed most dramatically at the Larsen A and B ice shelves, which collapsed rapidly in 1998 and 2002, when roughly 1250 square miles of ice fragmented and collapsed in little over a month.

The ice shelves are thinning because relatively warm water currents are eating away at them from below, but they’re also melting from the top as summer air temperatures rise. Both these effects thin and weaken the ice shelves and, as they do, the glaciers they are holding back flow faster to the sea and their margins retreat.

Using AUVs, the team were able to gather data on historic ice shelf fluctuations from the geological record on the Antarctic continental shelf.

“By examining landforms on the seafloor, we were able to make determinations about how the ice behaved in the past,” said Dowdeswell, who was chief scientist on the Weddell Sea Expedition. “We knew these features were there, but we’ve never been able to examine them in such great detail before.”

The team identified a series of delicate wave-like ridges on the seafloor, each only about one metre high and spaced 20 to 25 metres apart, dating to the end of the last great deglaciation of the Antarctic continental shelf, roughly 12,000 years ago. The researchers have interpreted these ridges as formed at what was formerly the grounding line – the zone where grounded ice sheet begins to float as an ice shelf.

The researchers inferred that these small ridges were caused by the ice moving up and down with the tides, squeezing the sediment into well-preserved geological patterns, looking a little like the rungs of a ladder, as the ice retreated. Assuming a standard 12-hour cycle between high and low tide, and measuring the distance between the ridges, the researchers were then able to determine how fast the ice was retreating at the end of the last Ice Age.

They calculated that the ice was retreating as much as 40 to 50 metres per day during this period, a rate that equates to more than 10 kilometres per year. In comparison, modern satellite images show that even the fastest-retreating grounding lines in Antarctica today, for example in Pine Island Bay, are much slower than these geological observations, at only about 1.6 kilometres per year.

“The deep marine environment is actually quite quiet offshore of Antarctica, allowing features such as these to be well-preserved through time on the seafloor,” said Dowdeswell. “We now know that the ice is capable of retreating at speeds far higher than what we see today. Should climate change continue to weaken the ice shelves in the coming decades, we could see similar rates of retreat, with profound implications for global sea-level rise.”

The research was funded in part by the Flotilla Foundation and Marine Archaeology Consultants Switzerland.

Reference:
J. A. Dowdeswell et al. ‘Delicate seafloor landforms reveal past Antarctic grounding-line retreat of kilometers per year.’ Science (2020). DOI: 10.1126/science.aaz3059

The researchers, from the University of Cambridge in collaboration with domestic appliance company Beko, used machine learning to train the robot to account for highly subjective matters of taste. The results are reported in the journal IEEE Robotics and Automation Letters, and will be available online as part of the virtual IEEE International Conference on Robotics and Automation (ICRA 2020).

A robot that can cook has been an aspiration of sci-fi authors, futurists, and scientists for decades. As artificial intelligence techniques have advanced, commercial companies have built prototype robot chefs, although none of these are currently commercially available, and they lag well behind their human counterparts in terms of skill.

“Cooking is a really interesting problem for roboticists, as humans can never be totally objective when it comes to food, so how do we as scientists assess whether the robot has done a good job?” said Dr Fumiya Iida from Cambridge’s Department of Engineering, who led the research.

Teaching a robot to prepare and cook food is a challenging task, since it must deal with complex problems in robot manipulation, computer vision, sensing and human-robot interaction, and produce a consistent end product.

In addition, taste differs from person to person – cooking is a qualitative task, while robots generally excel at quantitative tasks. Since taste is not universal, universal solutions don’t exist. Unlike other optimisation problems, special tools need to be developed for robots to prepare food.

Other research groups have trained robots to make cookies, pancakes and even pizza, but these robot chefs have not been optimised for the many subjective variables involved in cooking.

Egg dishes, omelettes in particular, have long been considered a test of culinary skill. A popular piece of French culinary mythology states that each of the one hundred pleats in a chef’s hat represents a different way to cook an egg, although the exact origin of this adage is unknown.

“An omelette is one of those dishes that is easy to make, but difficult to make well,” said Iida. “We thought it would be an ideal test to improve the abilities of a robot chef, and optimise for taste, texture, smell and appearance.”

In partnership with Beko, Iida and his colleagues trained their robot chef to prepare an omelette, from cracking the eggs through to plating the finished dish. The work was performed in Cambridge’s Department of Engineering, using a test kitchen supplied by Beko plc and Symphony Group.

The machine learning technique developed by Iida’s team makes use of a statistical tool, called Bayesian Inference, to squeeze out as much information as possible from the limited amount of data samples, which was necessary to avoid over-stuffing the human tasters with omelettes.

“Another challenge we faced was the subjectivity of human sense of taste - humans aren’t very good at giving absolute measures, and usually give relative ones when it comes to taste,” said Iida. “So we needed to tweak the machine learning algorithm - the so-called batch algorithm - so that human tasters could give information based on comparative evaluations, rather than sequential ones.”

But how did the robot measure up as a chef? “The omelettes, in general, tasted great – much better than expected!” said Iida.

The results show that machine learning can be used to obtain quantifiable improvements in food optimisation. Additionally, such an approach can be easily extended to multiple robotic chefs. Further studies have to be conducted to investigate other optimisation techniques and their viability.

“Beko is passionate about designing the kitchen of the future and believes robotics applications such as this will play a crucial part. We are very happy to be collaborating with Dr Iida on this important topic,” said Dr Graham Anderson, the industrial project supervisor from Beko’s Cambridge R&D Centre.

Reference:
Kai Junge et al. ‘Improving Robotic Cooking using Batch Bayesian Optimization.’ IEEE Robotics and Automation Letters (2020). DOI: 10.1109/LRA.2020.2965418

The first analysis of a new point-of-care 'nucleic acid test' for SARS-CoV-2 in a UK hospital setting shows these machines dramatically reduce time spent on COVID-19 ‘holding’ wards – allowing patients to be treated or discharged far quicker than with current lab testing set-ups.

The rapid diagnostic capability of SAMBA II devices – an average of 2.6 hours compared with 26.4 hours for standard lab tests – led to an increased availability of ‘isolation rooms’ needed for infected patients, as well as fewer hospital bay closures.

University of Cambridge researchers behind the new study, currently a pre-print and awaiting peer-review, say that the time and hospital capacity spared by these devices will be “critical as we move towards autumn and winter”.

The SAMBA II machine was developed by a University spinout company, Diagnostics in the Real World, and deployed for trials in Addenbrooke’s Hospital, part of Cambridge University Hospitals NHS Foundation Trust (CUH).

“The backlog of routine operations and screenings as a result of the pandemic is a huge issue, and must be resolved ahead of winter, when the NHS will face even more pressure from other infections like norovirus and influenza,” said study lead author Professor Ravi Gupta.

“Rapidly testing admissions for SARS-CoV-2 at the point of care is essential for reducing COVID-19 transmission in hospitals, speeding up access to urgent care and allowing safe discharge to care homes. It could make all the difference in a few months’ time.”

“Use of point-of-care testing would speed up the identification of patients for COVID-19 clinical trials, and receiving an experimental treatment a day earlier could make a clinical difference.”

“Hospitals across the UK, as well as care homes and prisons, could benefit from SAMBA II devices,” said Gupta, from the Cambridge Institute of Therapeutic Immunology & Infectious Diseases (CITIID). “Given the technological capital of the UK we should not be falling so short on rapid point-of-care testing.”

Standard tests are sent for analysis in central laboratories, where backlogs can see delays of two days or more. SAMBA machines can produce a diagnosis in as little as 90 minutes.

Dr Helen Lee, CEO of Diagnostics in the Real World, developed the SAMBA II technology while at Cambridge’s Department of Haematology. The chemistry behind the machines has been used for on-the-spot HIV diagnostics across Africa.

The devices search for tiny traces of virus genetic code, and are extremely sensitive in the detection of active infections. Once nose and throat swabs have been loaded into a SAMBA machine, the process is fully automated, making them easy to use.

The initial ‘COVIDx’ clinical study led by Gupta at Addenbrooke’s with 149 participants found SAMBA II had 96.9% sensitivity (accurate identification of positive cases) and 99.1% specificity (accurate identification of negative cases) compared to the standard lab test. It was also around 24 hours faster. 

The success of the COVIDx study saw the hospital switch nearly all of its SARS-CoV-2 testing from the standard lab ‘RT-PCR’ tests over to the use of SAMBA II during May: an opportunity for a “real-world” comparison and its effect on hospital functioning.

Gupta and colleagues compared data from the electronic patient records of all those who had in-hospital tests done in the ten days before and then after the switch to SAMBA devices at CUH.

The researchers found that the average length of time patients had to spend on a COVID019 ‘holding’ ward before they could be discharged or progress with treatment almost halved: dropping from 58.5 hours to just 30 hours.

They also saw a fall in use of the single-occupancy ‘isolation’ rooms in which COVID-19 patients are ideally treated – from 30.8% to 21.2% after SAMBA’s introduction, as patients with symptoms were shown to be COVID-19 negative.

In fact, the researchers say the testing devices prevented 11 ward closures in the ten days after implementation. “Keeping surgical bays open means fewer cancelled operations, speeding up access to life-saving clinical intervention,” said Gupta.

The majority of those tested using SAMBA II during the first ten days of hospital-wide use were admissions to the Emergency Department (ED). The remainder included pre-op screenings and discharges to nursing homes.

Some 96% of the SAMBA testing on pre-operative patients increased speed of “surgical intervention”, including kidney and liver transplants. The tests also allowed earlier discharge to nursing homes or into supported living in 79% of those cases, with the remainder delayed by “systematic issues” not tests themselves.

Dr Dami Collier, who coordinated and analysed COVIDx, said: “Our research demonstrates that point-of-care testing with SAMBA II machines is not only reliable, accurate and much faster, but that the diagnostic speed leads to significant real-world improvements for patient care and safety.”

CUH Medical Director, Dr Ashley Shaw, said: “Point of care testing has been hugely beneficial in enabling our clinical teams to make well-informed and timely decisions, keeping patients and staff as safe as possible throughout this difficult period.”

This work was supported by Wellcome, the Addenbrooke’s Charitable Trust, and the National Institute of Health Research (NIHR) Cambridge Biomedical Research Centre.
 

How you can support Cambridge's COVID-19 research effort

Donate to support COVID-19 research at Cambridge

That’s the warning from a team of researchers including Dr Ben Collier from the Cambridge Cybercrime Centre, part of Cambridge's Department of Computer Science and Technology.

The researchers have been analysing data collected by the Centre from underground forums, chat channels and marketplaces used by cybercrime communities. And in a briefing paper they have just written for Police Scotland, they say it indicates that the social changes put in place in response to the coronavirus pandemic 'have stimulated… the cybercrime economy.'

Some of the cybercrimes taking place are new. For example, early in the lockdown, some scammers sent fake texts, purporting to come from the UK's HM Revenue & Customs, telling recipients they were going to be fined £250 for leaving their homes more than once a day.​

And the researchers are also concerned that the rollout of the prospective NHS contact-tracing app has the potential to generate clear risks for those vulnerable to fraud. They warn that such people may be conned into handing over sensitive personal information by fake apps or scam texts purporting to be from the NHS.

“We’re also seeing some general repurposing of existing cybercrime,” said Collier. “For example, there have long been fake online shops, but now instead of selling clothes, they are selling face masks or bogus ‘cures’ for the coronavirus.”

And meanwhile, there has been a general rise in the levels of cybercrime.  The Cambridge Cybercrime Centre has tracked a three-fold increase in ‘denial of service’ attacks from around 12,000 per day to close to 30,000 attacks per day. These attacks – which can be purchased for small amounts of money from specialised online services – can be used to knock others offline, often opponents in online games.

Such attacks, the report says, have serious implications beyond being a nuisance for gamers, as many of these children and young people will be sharing internet connections with siblings engaged in online or blended learning and parents working from home.

We are vulnerable to such risks, Collier and his colleagues say, because we are spending much more time online as we work, or school our children, from home. And it is partly happening because “many internet users, including adolescents and young adults, are currently confined to home with no school or work for much of the day. The increased boredom they feel may well be a key driver of online petty crime.”

“Anxiety over serious economic problems – such as job losses and business closures – may be prompting some people to step up existing harmful online activity as a means of generating income,” said Collier. 

In their paper, the research team – Dr Collier, Dr Shane Horgan from Edinburgh Napier University, Dr Richard Jones from the University of Edinburgh and Dr Lynsay Shepherd from Abertay University – also voice their concerns about the potential for a steep rise in the volume of other online harms. These include online bullying, stalking and harassment of minority groups and victims of domestic abuse.

Their paper is a rapid response briefing aimed at offering guidance on the policing of cybercrime to Police Scotland. But its findings have relevance across the UK.

It says that while the UK has a sophisticated cybersecurity apparatus particularly at the national level, it currently lacks sufficient capability at the local level to police a significant increase in ‘volume’ cybercrime offences.

And it recommends that with levels of such crimes increasing, police forces need to engage more with their local communities and work with them on measures to prevent such crimes.

The paper also recommends that police forces, including Police Scotland, immediately undertake a wide-ranging review of their cybercrime policing and prevention practices and capabilities to assess their current adequacy and potential future resilience in the event that the number of cybercrime offences increases significantly in the near future.

The implications of the COVID-19 pandemic for cybercrime policing in Scotland: A rapid review of the evidence and future considerations’ is published by the Scottish Institute for Policing Research.

An open-source artificial intelligence (AI) tool, combining chest imaging data with laboratory and clinical data, is being developed by Cambridge researchers to support the rapid diagnosis and triaging of patients with COVID-19 in the UK.

The team, led by Professors Carola-Bibiane Schönlieb and Evis Sala, brings together expertise in AI for imaging with expertise in radiology and clinical applications from Addenbrooke’s and Papworth Hospitals, as well as collaborators from the UK, China, Austria and Italy, to develop a prediction model that can rapidly and reliably diagnose and suggest a prognosis to doctors.

Reverse-transcription polymerase chain reaction (RT-PCR) tests are currently the most common tool used to diagnose COVID-19, but they are only up to 70% sensitive, meaning there are up to 30% false negatives.

While chest X-rays and CT scans provide valuable diagnostic and monitoring information that can complement laboratory and clinical data, it is a complex task typically done by radiologists, whose expertise is often in high demand. Fast and accurate diagnosis of patients in order to limit disease spread, together with the rapid determination of whether a patient is likely to recover, require intensive care unit (ICU) admission, or intensive ventilation, is key to allocating resources and to improving patient outcomes.

“AI offers huge potential to support agile clinical decision making, ensuring patients receive the most appropriate support and leading to better patient outcomes,” said Sala, who is based at the Department of Radiology.

Recent studies have suggested that using AI could have a meaningful impact on the management of patients with COVID-19. AI tools such as deep learning can offer automated image analysis and integration with clinical data to help clinicians make more informed decisions for treatment.

However, good quality data and computing power are required to train and optimise predictive AI models and data availability is a major bottleneck when developing new systems. Coupled with this, the lack of standardisation of datasets makes it challenging to reuse existing AI tools in a different country than the one that it was trained for. Most current AI tools have been developed on small, locally collected datasets. Data that is being collected in hospitals all over the world varies in what is being collected and how the data is processed. Therefore, an effort for developing a widely applicable tool for COVID-19 hospital support must be open source so it can be adapted to different environments; be based on a serious data sharing and data curation, data cleaning and standardisation effort; and be developed with mathematical, statistical and engineering expertise to develop robust and translatable tools.

To address these challenges, the team from the Cambridge Centre for Mathematical Imaging in Healthcare (CMIH) is developing a flexible, open-source AI tool that could be used by hospitals worldwide. Drawing on their history of global research collaboration and expertise in data governance, the team is gathering datasets from Austria, China, Italy and the UK for their work. Data scientists and clinicians are working in close collaboration, following standard protocols to identify bias during development. “Rigorous mathematical models play a key role in mitigating bias and improving the efficacy of the prediction model as they follow universal rules with mathematical guarantees,” said Schönlieb.

Using deep learning approaches along with mathematical and statistical analysis methods, the new tool will be accompanied by a comprehensive algorithmic strategy that will allow fine-tuning for datasets with different characteristics and implementation in different countries. The team are hoping to launch the AIX-COV-NET tool within the next 12 to 18 months. The project has recently received funding from the EU-funded Innovative Medicines Initiative and Intel.

“Our team’s strength is the close dialogue we have between clinicians and data scientists, and the passion we all bring for advancing AI solutions for COVID-19,” said Schönlieb, from Cambridge's Department of Applied Mathematics and Theoretical Physics.

“AI offers huge potential to support agile clinical decision making, ensuring patients receive the most appropriate support and leading to better patient outcomes,” said Sala.

The core project team is comprised of data scientists and clinicians from across Cambridge and is led by Professor Carola-Bibiane Schönlieb, Director of the Centre for Mathematical Imaging in Healthcare, and Professor Evis Sala, Professor of Oncological Imaging, University of Cambridge & Honorary Consultant Radiologist, Addenbrooke’s Hospital, Cambridge University Hospitals NHS Foundation Trust. The team is supported by an AI and image analysis team, drawn from subject experts across Cambridge and around the world, a clinical team comprised of colleagues from hospitals in Cambridge, London and Vienna, and a support team based in the Faculty of Mathematics. Partner institutions include hospitals in Wuhan, China; Milan, Italy; and Madrid, Spain, and universities in Manchester, Vienna and London.

The University of Cambridge has an impressive record of achievement in multidisciplinary research and innovation. The CMIH is a collaboration between mathematics, engineering, physics and biomedical scientists and clinicians and is one of five centres to receive investment from the Engineering and Physical Sciences Research Council (EPSRC). A key aim of this partnership is the delivery of high quality, multidisciplinary research that will help overcome some of the major challenges facing the NHS.

How you can support Cambridge's COVID-19 research effort

Donate to support COVID-19 research at Cambridge

I’m a clinical research fellow and veterinary surgeon in the new Cambridge Institute for Therapeutic Immunology and Infectious Diseases (CITIID) on the Biomedical Campus. Over the past few months I have divided my time between CITIID, and volunteering in the Department of Virology as part of the COVID-19 Genomics Consortium.

My initial role in the COVID-19 pandemic was related to diagnostics. I gained experience of testing patient samples for viruses during the Ebola outbreak in 2015 in Sierra Leone, so when COVID-19 cases started rising in the UK I volunteered to help the Public Health England lab in Addenbrookes. From there I joined Professor Ian Goodfellow’s team working to sequence full genomes of the virus from patients across East Anglia. As case numbers are being brought under control I’ve been able to transition back to virus research, which aims to improve our understanding of coronavirus immunity.

My research usually focuses on the antibody response to viruses. This means it hasn’t been too large a leap to extend my research to include coronaviruses. We need to determine how coronavirus-specific antibodies are working, in order to find out what the ‘ideal’ antibody response to SARS-CoV-2 is. This will be valuable for development of effective vaccines and for identification of people that may be susceptible to repeat infections.  

As a veterinarian, I have also been closely following the news about COVID-19 in animals. There are many myths and misconceptions in this area, so I‘ve been actively engaged in reassuring veterinary professionals and the public about risks to pets. I’ve written a number of articles for The Conversation and the Naked Scientists on this in recent weeks (there’s still zero evidence of pets transmitting the virus to humans). With support from colleagues in CITIID, I have also established a new project for COVID-19 testing in animals. 

Development of safe, effective, and widely available vaccines is an incredible challenge facing scientists right now. Many different vaccine approaches are being studied, but we don’t yet know which is going to be successful. The number of trials currently underway for SARS-CoV-2 vaccines is beyond anything the vaccine field has previously seen.

I have always been keen to seen molecular biology translated for use in medical settings. There is now cross-talk between research institutes and the hospitals like never before, and the scientists and clinicians are collaborating at an impressive rate!

Many lessons have been learnt about our ability and capacity to test for viruses in the UK. The number of scientists wanting to volunteer to help with testing has been immense, so in future I hope logistics and organisation will be able to match this enthusiasm much quicker.

When the pandemic is over, I’m looking forward to travelling anywhere outside of Cambridge! Much as I love this city, I really miss venturing further afield to see family, friends, and explore new places. I’d also really like to sit in a busy café with a good coffee!

Sarah Caddy is a Wellcome Trust Clinical Research Fellow in Viral Immunology at the Cambridge Institute for Therapeutic Immunology and Infectious Disease.

How you can support Cambridge’s COVID-19 research

Despite years of experience advising the government, Professor Neil Ferguson couldn’t have anticipated that his private life would become a matter of public scrutiny last month, essentially ending his formal relationship with government. No less surprising is the move from a former Government Chief Scientific Adviser, Professor Sir David King, to set up a so-called ‘Independent SAGE’ - laying bare the kinds of deliberations that would have taken place behind closed doors during his appointment.

The bottom line for high-profile scientists and scientific advisers is that the rules of the game have changed. They may have learned that discretion is highly valued by policymakers, and yet, calls for transparency continue to resound louder than ever. How are scientists dealing with these new circumstances? What are they learning?

In a recent paper, I argue that too little attention has been given to how experts learn to advise policymakers. Although there is no shortage of guidelines and fragments of wisdom for researchers who want to see their work (or the work of colleagues) inform policymaking, scientific advice to governments is largely a case of learning on the job.

In their role as scientific advisers, experts learn what is and isn’t appropriate behaviour, what is and isn’t politically acceptable, and to draw the line where the science ‘ends’ and the politics ‘begin’. Scientific advice is a tricky balancing act between making expert judgments on the best available evidence and calibrating those judgments to the politics of the issues at hand. Like a tightrope walker, the scientific adviser has to learn to get the balance just right.

The journey from full-time academic to part-time scientific adviser can be a transformative one. Researchers might initially set out with expectations of how scientists and policymakers interact and have had to revisit those expectations in view of their various encounters and experiences. While their learning may not always be transformative, I suggest that it is always necessarily situated: different organisations and environments will influence and shape their learning in different ways. This includes discussions with peers on scientific advisory committees, for example. How and what advisers learn, then, is never quite divorced from where they learn.

So why should we care? Taking the long view, I see three reasons why we might want to put advisers’ learning under the microscope:

1. Compiling the know-how of experienced advisers can be helpful for less experienced or early-career researchers who wish to engage with policymakers. Because there is no universal roadmap for success, I think we should focus on coming up with some ‘warning signs’ - as opposed to ‘direction signs’ - by identifying and communicating common pitfalls, for instance.
2. Given the situated nature of their own learning, advisers can directly contribute to the institutional learning and memory of the science-policy organisations they are part of. Involving committee members in decision-making can help prevent needless reinvention of the wheel and improve organisational reforms, leading to more sustainable change.
3. As both academics and policy advisers, scientific advisers are particularly well-placed to understand how academic research informs (or fails to inform) policymaking, as well as how the scientific community works and is governed. Therefore, they are knowledgeable not only about science for policy, but also about policy for science. For those reasons, I think that research funding organisations – such as UKRI research councils - should more systematically consult experienced science advisers in the formulation of their policies, especially in relation to research impact. For instance, improved impact evaluation frameworks would have positive trickle-down effects on the wider academic community, especially for early-career researchers who tend to base their understanding of impact in large part on the existing guidelines for grant applications or job descriptions.

The core message is that as the nature of both science and policymaking continues to change, the learning experiences of expert advisers is an abundant resource that has yet to be tapped into. This has become all the more evident with COVID-19, as scientific advisers’ learning curves are likely to be steep. In the aftermath of the pandemic, we’ll need an evaluation of ‘what happened’ and ‘what went wrong’.

For the whole picture, we can’t just rely on the loudest or the most visible voices. We’ll need to turn to those scientific advisers whose stories go largely untold. Importantly, we’ll need to understand why the acquired skillset of scientific advisers may not be suited for crisis situations. Only then can we ensure that lessons are learned and that our networks of science advice are prepared for future emergencies.

Plastic has become a malevolent symbol of our wasteful society. It’s also cheap, durable, flexible, waterproof, versatile, lightweight, protective and hygienic.

During the coronavirus pandemic, plastic visors, goggles, gloves and aprons have been fundamental for protecting healthcare workers from the virus. But what about the effects on the environment of throwing away huge numbers of single-use medical protection equipment? How are we to balance our need for plastic with protecting the environment?

Released on 5 June 2020, World Environment Day, this new film considers how society might ‘reset the clock’ when it comes to living better with a vital material. We hear how Cambridge University's Cambridge Creative Circular Plastics Centre (CirPlas) aims to eliminate plastic waste by combining blue-sky thinking with practical measures – from turning waste plastic into hydrogen fuel, to manufacturing more sustainable materials, to driving innovations in plastic recycling in a circular economy.

“Plastic is an example of how we must find ways to use resources without irreversibly changing the planet for future generations,” adds Professor Erwin Reisner, who leads CirPlas, which is funded by UK Research and Innovation.

Explore more:

Find further information on CirPlas

Read more about our research on plastic

Visit our spotlight on Sustainable Earth

Real-time tracking of a pandemic, as data accumulate over time, is an essential component of a public health response to a new outbreak. A team of statistical modellers at the Unit are working with Public Health England (PHE) to regularly ‘nowcast’ and forecast COVID-19 infections and deaths. This information feeds directly to SAGE sub-group, Scientific Pandemic Influenza sub-group on Modelling (SPI-M) and to regional PHE teams.

The work uses a transmission model, data on daily COVID-19 confirmed deaths from PHE, and published information on the risk of dying and the time from infection to death, to reconstruct the number of new COVID-19 infections over time, estimate a measure of ongoing transmission, and predict the number of new COVID-19 deaths in different regions and age groups.

In the latest findings, published today, the researchers say:

• There are an estimated 17,000 new infections arising each day across England
• The number of deaths each day is likely to fall to between 100–250 by mid-June
• The R number is below 1 in all regions of England with the exception of the North West and the South West
• In the South West, although R is around 1, the numbers of new infections occurring in the region on a daily basis is relatively low
• There is some evidence that R has risen in all regions, probably due to increasing mobility and mixing between households and in public and workplace settings
• This increase in R will lead to a slowdown in the decrease in new infections and deaths
• The increases in the regional R numbers may result in the decline in the national death rate being arrested by mid-June

For further details, see Nowcasting and Forecasting of COVID-19.

The code, script and documentation of the 16 March report, which is available on Github, was subject to an independent review led by Dr Stephen Eglen, from Cambridge’s Department of Applied Mathematics and Theoretical Physics.

Eglen co-founded Codecheck last year to help evaluate the computer programs behind scientific studies. Researchers provide their code and data to Codecheck, who run the code independently to ensure the work can be reproduced.

Last week, Codecheck certified the reproducibility of arguably the most talked-about computational model of the COVID-19 pandemic, that of the Imperial College group led by Professor Neil Ferguson. The model suggested that there could be up to half a million deaths in the UK if no measures were taken to slow the spread of the virus, and has been cited as one of the main reasons that lockdown went into effect soon after. However, the Imperial group did not immediately make their code publicly available.

Codecheck.org.uk provided an independent review of the replication of key findings from Report 9 using CovidSim reimplementation. The process matches domain expertise and technical skills, taking place as an open peer review. The reviewer conducts the codecheck and submits the resulting certificate as part of their review.

The results confirm that the key finding of Report 9 - on the impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand - are reproducible. Eglen did not review the epidemiology that went into the Imperial model, however.

In his analysis, Dr Eglen said: “Each run generated a tab-delimited file in the output folder. Two R scripts provided by Prof Ferguson were used to summarise these runs into two summary files... These files were compared against the values generated by Prof Ferguson...The results were found to be identical. Inserting my results into his Excel spreadsheet generated the same pivot tables.”

The codecheck found that: “Small variations (mostly under 5%) in the numbers were observed between Report 9 and our runs.” The codecheck confirmed the trends and findings of the original report.

Building in part on code originally developed, published and peer-reviewed in 2005 and 2006, the code used for Report 9 continues to be actively developed to allow examination of the wider range of control policies now being deployed as countries relax lockdown. The Imperial team is sharing the code to enhance transparency and to allow others to contribute and make use of the simulation.

Refactoring the code has allowed changes to be made more quickly and reliably, including incorporating new data that has become available as the pandemic has progressed.

In addition to the features presented in Imperial Report 9, further strategies can now be examined such as testing and contact tracing, which was not a UK policy option in March.

Users also now have the ability to vary intensity of interventions over time and to calibrate the model to country-specific epidemic data.

Adapted from a piece originally published on the Imperial College London website

How you can support Cambridge's COVID-19 research effort

Donate to support COVID-19 research at Cambridge

A team of engineers, radiologists and physicians, led by the University of Cambridge, developed the algorithm, which builds a three-dimensional model of an individual’s knee joint in order to map where arthritis is affecting the knee. It then automatically creates ‘change maps’ which not only tell researchers whether there have been significant changes during the study but allow them to locate exactly where these are.

There are few effective treatments for arthritis, and the technique could be a considerable boost to efforts to develop and monitor new therapies for the condition. The results are reported in the Journal of Magnetic Resonance Imaging.

Osteoarthritis is the most common form of arthritis in the UK. It develops when the articular cartilage that coats the ends of bones and allows them to glide smoothly over each other at joints, is worn down, resulting in painful, immobile joints. Currently, there is no recognised cure and the only definitive treatment is surgery for artificial joint replacement.

Osteoarthritis is normally identified on an X-ray by a narrowing of the space between the bones of the joint due to a loss of cartilage. However, X-rays do not have enough sensitivity to detect subtle changes in the joint over time.

“We don’t have a good way of detecting these tiny changes in the joint over time in order to see if treatments are having any effect,” said Dr James MacKay from Cambridge’s Department of Radiology, and the study’s lead author. “In addition, if we’re able to detect the early signs of cartilage breakdown in joints, it will help us understand the disease better, which could lead to new treatments for this painful condition.”

The current study builds on earlier work from the same team, who developed an algorithm to monitor subtle changes in arthritic joints in CT scans. Now, they are using similar techniques for MRI, which provides more complete information about the composition of tissue – not just information about the thickness of cartilage or bone.

MRI is already widely used to diagnose joint problems, including arthritis, but manually labelling each image is time-consuming, and may be less accurate than automated or semi-automated techniques when detecting small changes over a period of months or years.

“Thanks to the engineering expertise of our team, we now have a better way of looking at the joint,” said MacKay.

The technique MacKay and his colleagues from Cambridge’s Department of Engineering developed, called 3D cartilage surface mapping (3D-CaSM), was able to pick up changes over a period of six months that weren’t detected using standard X-ray or MRI techniques.

The researchers tested their algorithm on knee joints from bodies that had been donated for medical research, and a further study with human participants between 40 and 60 years old. All of the participants suffered from knee pain, but were considered too young for a knee replacement. Their joints were then compared with people of a similar age with no joint pain.

“There’s a certain degree of deterioration of the joint that happens as a normal part of aging, but we wanted to make sure that the changes we were detecting were caused by arthritis,” said MacKay. “The increased sensitivity that 3D-CaSM provides allows us to make this distinction, which we hope will make it a valuable tool for testing the effectiveness of new therapies.”

The software is freely available to download and can be added to existing systems. MacKay says that the algorithm can easily be added to existing workflows and that the training process for radiologists is short and straightforward. 

As part of a separate study funded by the European Union, the researchers will also be using the algorithm to test whether it can predict which patients will need a knee replacement, by detecting early signs of arthritis.

Reference:
James W. MacKay et al. ‘Three-dimensional Surface-based Analysis of Cartilage MRI data in Knee Osteoarthritis: Validation and Initial Clinical Application.’ Journal of Magnetic Resonance Imaging (2020). DOI: 10.1002/jmri.27193

Population-wide use of facemasks keeps the coronavirus ‘reproduction number’ under 1.0, and prevents further waves of the virus when combined with lockdowns, a modelling study led by the University of Cambridge suggests.

The research suggests that lockdowns alone will not stop the resurgence of SARS-CoV-2, and that even homemade masks with limited effectiveness can dramatically reduce transmission rates if worn by enough people, regardless of whether they show symptoms. 

The researchers call for information campaigns across wealthy and developing nations alike that appeal to our altruistic side: 'my facemask protects you, your facemask protects me'. The findings are published in the Proceedings of the Royal Society A.

“Our analyses support the immediate and universal adoption of facemasks by the public,” said lead author Dr Richard Stutt, part of a team that usually models the spread of crop diseases at Cambridge’s Department of Plant Sciences.

“If widespread facemask use by the public is combined with physical distancing and some lockdown, it may offer an acceptable way of managing the pandemic and reopening economic activity long before there is a working vaccine.”

Dr Renata Retkute, coauthor and Cambridge team member, said: “The UK government can help by issuing clear instructions on how to make and safely use homemade masks.”

“We have little to lose from the widespread adoption of facemasks, but the gains could be significant.”

The new coronavirus is transmitted through airborne droplets loaded with SARS-CoV-2 particles that get exhaled by infectious people, particularly when talking, coughing or sneezing.

For the latest study, Cambridge researchers worked to link the dynamics of spread between individuals with population-level models, to assess different scenarios of facemask adoption combined with periods of lockdown.

The modelling included stages of infection and transmission via surfaces as well as air. Researchers also considered negative aspects of mask use, such as increased face touching.

The reproduction or ‘R’ number – the number of people an infected individual passes the virus onto – needs to stay below 1.0 for the pandemic to slow.

The study found that if people wear masks whenever they are in public it is twice as effective at reducing ‘R’ than if masks are only worn after symptoms appear.

In all modelling scenarios, routine facemask use by 50% or more of the population reduced COVID-19 spread to an R less than 1.0, flattening future disease waves and allowing less-stringent lockdowns.

Viral spread reduced further as more people adopted masks when in public. 100% mask adoption combined with on/off lockdowns prevented any further disease resurgence for the 18 months required for a possible vaccine.   

The models suggest that – while the sooner the better – a policy of total facemask adoption can still prevent a second wave even if it isn’t instigated until 120 days after an epidemic begins (defined as the first 100 cases).

The team investigated the varying effectiveness of facemasks. Previous research shows that even homemade masks made from cotton t-shirts or dishcloths can prove 90% effective at preventing transmission.

The study suggests that an entire population wearing masks of just 75% effectiveness can bring a very high ‘R’ number of 4.0 – the UK was close to this before lockdown – all the way down to under 1.0, even without aid of lockdowns.

In fact, masks that only capture a mere 50% of exhaled droplets would still provide a 'population-level benefit', even if they quadrupled the wearer’s own contamination risk through frequent face touching and mask adjustment (a highly unlikely scenario).

The researchers point out that crude homemade masks primarily reduce disease spread by catching the wearer’s own virus particles, breathed directly into fabric, whereas inhaled air is often sucked in around the exposed sides of the mask.

“There is a common perception that wearing a facemask means you consider others a danger,” said Professor John Colvin, coauthor from the University of Greenwich. “In fact, by wearing a mask you are primarily protecting others from yourself.”

“Cultural and even political issues may stop people wearing facemasks, so the message needs to be clear: my mask protects you, your mask protects me.”

“In the UK, the approach to facemasks should go further than just public transport. The most effective way to restart daily life is to encourage everyone to wear some kind of mask whenever they are in public,” Colvin said.

Professor Chris Gilligan, coauthor from Cambridge’s Epidemiology and Modelling Group in the Department of Plant Sciences, added: “These messages will be vital if the disease takes hold in the developing world, where large numbers of people are resource poor, but homemade masks are a cheap and effective technology.”
 

How you can support Cambridge's COVID-19 research effort

Donate to support COVID-19 research at Cambridge

I work in various places, including Cambridge, Luxembourg, and Wuhan. I was in Wuhan for a month last November, and luckily left right before the outbreak started. In anticipation of the travel ban, I cancelled all my trips and went to Luxembourg, where my family lives.

Computational tools can be tremendously helpful in searching large datasets and extracting key information. I enjoy digging through the complexity inherent in biological and biomedical systems. I use mathematical models to generate and test new ideas, provide insights into biological mechanisms, and try to pinpoint the source of diseases. My normal research ranges from understanding the mechanisms of the circadian clocks in plants, to anticipating heart attacks in humans. I’m now using my expertise to help predict the impact of COVID-19 on our lives. 

I was approached by Luxembourg’s National COVID-19 Task Force and the government to predict how COVID-19 will spread in Luxembourg. Half of my group volunteered to help. For over two months we’ve been simulating different lockdown exit scenarios. We make short- and mid-term projections on the number of infected people, ICU occupancy and number of deaths. Luxembourg is a unique country: due to its small size, we can simulate how the virus spreads through the entire real work, household and school networks. This allows us to build epidemiological models at the individual level to test different exit strategies. 

The main objective of the exit strategy is to safely revive the economy. This means simulating a large number of alternative ways to progressively open key sectors of the economy, while keeping the number of newly infected people to a minimum. This information has been the basis on which the government of Luxembourg makes their decisions. Examples of our simulations can be found in the 25 May 2020 report. 

In another project, we’ve developed a model to predict COVID-19 patient survival with over 90% accuracy. With collaborators from Huazhong University of Science and Technology in Wuhan, we identified three key blood-borne biomarkers that help predict mortality 10 days in advance. We analysed blood samples from almost 500 COVID-19 patients from Wuhan and trained a machine learning algorithm to reliably predict those at high risk of death. The goal is to identify high-risk COVID-19 patients needing urgent care as early as possible. The work was recently published in the journal Nature Machine Intelligence.

Although we’ve been in lockdown for over two months we’re really only at the beginning of the pandemic. At the moment about 2% to 5% of the population is infected with COVID-19 globally. This means that 95% of the population is still susceptible. Heard immunity is only reached after at least two-thirds of the population has been infected. Reaching it by infecting each other comes at an unacceptably high loss of life, even if the health care system does not saturate. Heard immunity should only be reached with a vaccine. 

Patience is therefore one of the biggest challenges we face today. Governments must continue to remind everyone of the risks associated with the virus, and educate people to be responsible while easing confinement measures. At the same time, we need to be creative with safe social interactions to safeguard the mental health of children, seniors, and the most vulnerable. After COVID-19 we should be better prepared to face another viral threat. We’re now familiar with face masks, social distancing and contact tracing, which are key steps in slowing the spread of a virus.

It has been impressive to see our community, the country and the world came together to tackle this problem. Academics put their research aside and started collaborating with a wide range of specialists ranging from virologists, medical doctors, epidemiologists, engineers, statisticians, physicists, social scientists, economists, and others. No-one can do this alone, and we have been helping each other understand this pandemic, putting aside competition and sharing information. 

When the pandemic is over, I’m looking forward to having a pint of beer with my friends, going on holiday with my family, and getting back to my core research. In that order!

Jorge Goncalves holds positions as Principal Research Associate in Information Engineering in the University of Cambridge’s Department of Plant Sciences, and Head of the Systems Control group at the University of Luxembourg’s Centre for Systems Biomedicine.

How you can support Cambridge’s COVID-19 research

The COV002 trial, developed by the University of Oxford, aims to assess how well people across a broad range of ages could be protected from COVID-19 using a new vaccine called ChAdOx1 nCoV-19. It will also provide valuable information on safety of the vaccine and its ability to generate good immune responses against the virus.

Cambridgeshire and Peterborough NHS Foundation Trust (CPFT), Cambridge University Hospitals NHS Foundation Trust and Royal Papworth Hospital NHS Foundation Trust are recruiting healthy staff aged between 18-55 years old who have not been infected with coronavirus but have regular face-to-face contact with COVID-19 patients, to take part in the trial.

Eligible participants will be randomised to receive one dose of either the trial vaccine (ChAdOx1 nCoV-19) or a licensed meningitis vaccine (MenACWY) that will be used as a ‘control’ for comparison. Following vaccination, participants will be followed up over 12 months.

Dr Estée Török from the Department of Medicine at the University of Cambridge and Principal Investigator at Cambridge University Hospitals NHS Foundation Trust, said: “Developing an effective vaccine is key to controlling the COVID-19 pandemic. We are delighted to be working with CPFT and Royal Papworth on this UK national priority vaccine trial. We are looking for healthy volunteers at high risk of COVID-19 infection at CUH to participate in this study and are most grateful to them for doing so.”

Dr Ben Underwood, Deputy Medical Director and Principal Investigator (study lead) at CPFT said: “We are grateful to all our staff for their brilliant response to the coronavirus pandemic.  Our research teams are playing a vital role in international efforts to secure a vaccine, which we hope will protect those most at risk, save more lives and minimise the disruption caused by the virus. Thank you to all volunteers who take part and make clinical trials possible.”

Dr Robert Rintoul, Director, Papworth Trials Unit Collaboration, and Reader in Thoracic Oncology at the Department of Oncology, University of Cambridge, said: “We at Royal Papworth Hospital are proud to be supporting research into possible vaccines and treatments for COVID-19. I would like to thank our staff members who have chosen to participate in this important public health project.”

Adapted from a press release by Cambridge University Health Partners.

In a study aimed at understanding the effect of therapeutic drugs on the brains of people living with Huntington’s disease, researchers used electroencephalography (EEG) to measure immediate changes in the animals’ brain waves once ketamine - an anaesthetic and pain relief drug - was administered. Low frequency activity dominated while the sheep were asleep. When the drug wore off and the sheep regained consciousness, the researchers were surprised to see the brain activity start switching between high and low frequency oscillations. The bursts of different frequency were irregular at first, but became regular within a few minutes.

“As the sheep came round from the ketamine, their brain activity was really unusual,” said Professor Jenny Morton at the University of Cambridge’s Department of Physiology, Development and Neuroscience, who led the research. “The timing of the unusual patterns of sheep brain activity corresponded to the time when human users report feeling their brain has disconnected from their body.”

She added:  “It’s likely that the brain oscillations caused by the drug may prevent information from the outside world being processed normally,”

The findings arose as part of a larger research project into Huntington’s disease, a condition that stops the brain working properly. The team want to understand why human patients respond differently to various drugs if they carry the gene for this disease. Sheep were used because they are recognised as a suitable pre-clinical model of disorders of the human nervous system, including Huntington’s disease. 

Six of the sheep were given a single higher dose of ketamine, 24mg/kg. This is at the high end of the anaesthetic range. Initially, the same response was seen as with a lower dose. But within two minutes of administering the drug, the brain activity of five of these six sheep stopped completely, one of them for several minutes - a phenomenon that has never been seen before. 

“This wasn’t just reduced brain activity. After the high dose of ketamine the brains of these sheep completely stopped. We’ve never seen that before,” said Morton. Although the anaesthetised sheep looked as though they were asleep, their brains had switched off. “A few minutes later their brains were functioning normally again - it was as though they had just been switched off and on.” 

The researchers think that this pause in brain activity may correspond to what ketamine abusers describe as the ‘K-hole’ - a state of oblivion likened to a near-death experience, which is followed by a feeling of great serenity. The study is published today in the journal Scientific Reports.

Ketamine abusers are known to take doses many times higher than those given to the sheep in this research. It is also likely that progressively higher doses have to be taken to get the same effect. The researchers say that such high doses can cause liver damage, may stop the heart, and be fatal.

To conduct the experiment sheep were put into veterinary slings, which are commonly used to keep animals safe during veterinary procedures. Different doses of ketamine were given to 12 sheep and their brain activity recorded with EEG.

Ketamine was chosen for the study because it is widely used as a safe anaesthetic and pain-relief drug for treating large animals including dogs, horses and sheep. It is also used medically, and is known as a ‘dissociative anaesthetic’ because patients can appear awake and move around, but they don’t feel pain or process information normally - many report feeling as though their mind has separated from their body. 

At lower doses ketamine has a pain-relieving effect, and its use in adult humans is mainly restricted to field situations such as frontline pain-relief for injured soldiers or victims of road traffic accidents.

“Our purpose wasn't really to look at the effects of ketamine, but to use it as a tool to probe the brain activity in sheep with and without the Huntington’s disease gene,” said Morton. “But our surprising findings could help explain how ketamine works. If it disrupts the networks between different regions of the brain, this could make it a useful tool to study how brain networks function - both in the healthy brain and in neurological diseases like Huntington’s disease and schizophrenia.”

Ketamine has recently been proposed as a new treatment for depression and post-traumatic stress disorder. Beyond its anaesthetic actions, however, very little is known about its effects on brain function.

“We think of anaesthetic drugs as just slowing everything down. That's what it looks like from the outside: the animals basically go to sleep and are unresponsive, and then they wake up very quickly. But when we looked at the brain activity, it seems to be a much more dynamic process,” said Morton. 

This research was funded by CHDI Inc. It was reviewed and approved by the Ethics Committee of the University of Cambridge.

Reference
Nicol, A.U. & Morton, A.J. ‘Characteristic patterns of EEG oscillations in sheep (Ovis aries) induced by ketamine may explain the psychotropic effects seen in humans.’ Scientific Reports, June 2020. DOI: 1038/s41598-020-66023-8

Using data science techniques, the team revealed that the NHS trust in which a COVID-19 patient ended up in intensive care is as important, in terms of the risk of death, as the strongest patient-specific risk factors such as older age, immunosuppression or chronic heart/kidney disease. In the worst case, COVID-19 patients in the intensive care unit (ICU) of a particular NHS trust were over four times as likely to die in a given time period than COVID-19 patients in an average trust’s ICU.

From the earliest days of the coronavirus pandemic, clinicians and scientists have been deciphering the risk factors that make someone with COVID-19 more likely to die. The uncovering of determinants of risk has allowed doctors to focus resources on the most vulnerable patients and has proved important in planning for the surge in demand for intensive care units created by the pandemic. It has also informed the public of which groups should take greater measures to shield or socially distance themselves. The new study is the first to reveal the extent to which ICU-patient location is a factor.

 “COVID-19 has stretched most ICUs well beyond their normal capacity and necessitated them finding additional space, equipment and skilled staff – in an already stretched NHS – to deal with demand for highly specialist life-supporting therapies,” says Dr Ercole. “It is possible that some hospitals found this harder either because they didn’t have time to react or the necessary resources. It is crucial to understand the reasons for these between-centre differences as we plan our response to similar situations in the future: how and where to build capacity, and how to use what we have most effectively.”

The peer-reviewed paper – “Between-centre differences for COVID-19 ICU mortality from early data in England” – has been accepted for publication in Intensive Care Medicine. A preprint of the study, posted prior to the completion of peer-review, is available online.

The analysis was carried out on anonymised data from the COVID-19 Hospitalisation in England Surveillance System (CHESS) dataset, supplied by Public Health England. The data were anonymised not only in terms of the patients but also in terms of the NHS trusts. The data covered 8 February to 22 May, during which there were 5062 ICU cases in 94 NHS trusts across England, with 1547 patient deaths and 1618 discharges from ICU.

The researchers call for urgent “comparative effectiveness research” to get to the bottom of these marked differences between NHS trusts. Knowledge gained in this direction could inform how ICUs are optimised and improve best practice in dealing with surges in COVID-19 cases in England, and perhaps beyond.

The team, from the University of Cambridge and the University of Dundee, developed a new context-aware method that reduces this communication gap by eliminating between 50% and 96% of the keystrokes the person has to type to communicate.

The system is specifically tailed for nonverbal people and uses a range of context ‘clues’ – such as the user’s location, the time of day or the identity of the user’s speaking partner – to assist in suggesting sentences that are the most relevant for the user.

Nonverbal people with motor disabilities often use a computer with speech output to communicate with others. However, even without a physical disability that affects the typing process, these communication aids are too slow and error-prone for meaningful conversation: typical typing rates are between five and 20 words per minute, while a typical speaking rate is in the range of 100 to 140 words per minute.

“This difference in communication rates is referred to as the communication gap,” said Professor Per Ola Kristensson from Cambridge’s Department of Engineering, the study’s lead author. “The gap is typically between 80 and 135 words per minute and affects the quality of everyday interactions for people who rely on computers to communicate.”

The method developed by Kristensson and his colleagues uses artificial intelligence to allow a user to quickly retrieve sentences they have typed in the past. Prior research has shown that people who rely on speech synthesis, just like everyone else, tend to reuse many of the same phrases and sentences in everyday conversation. However, retrieving these phrases and sentences is a time-consuming process for users of existing speech synthesis technologies, further slowing down the flow of conversation.

In the new system, as the person is typing, the system uses information retrieval algorithms to automatically retrieve the most relevant previous sentences based on the text typed and the context the conversation the person is involved in. Context includes information about the conversation such as the location, time of day, and automatic identification of the speaking partner’s face. The other speaker is identified using a computer vision algorithm trained to recognise human faces from a front-mounted camera.

The system was developed using design engineering methods typically used for jet engines or medical devices. The researchers first identified the critical functions of the system, such as the word auto-complete function and the sentence retrieval function. After these functions had been identified, the researchers simulated a nonverbal person typing a large set of sentences from a sentence set representative of the type of text a nonverbal person would like to communicate.

This analysis allowed the researchers to understand the best method for retrieving sentences and the impact of a range of parameters on performance, such as the accuracy of word-auto complete and the impact of using many context tags. For example, this analysis revealed that only two reasonably accurate context tags are required to provide the majority of the gain. Word-auto complete provides a positive contribution but is not essential for realising the majority of the gain. The sentences are retrieved using information retrieval algorithms, similar to web search. Context tags are added to the words the user types to form a query.

The study is the first to integrate context-aware information retrieval with speech-generating devices for people with motor disabilities, demonstrating how context-sensitive artificial intelligence can improve the lives of people with motor disabilities.

“This method gives us hope for more innovative AI-infused systems to help people with motor disabilities to communicate in the future,” said Kristensson. “We’ve shown it’s possible to reduce the opportunity cost of not doing innovative research with AI-infused user interfaces that challenge traditional user interface design mantra and processes.”

The research paper was published at CHI 2020.

The research was funded by the Engineering and Physical Sciences Research Council.

Reference:
Kristensson, P.O., Lilley, J., Black, R. and Waller, A. ‘A design engineering approach for quantitatively exploring context-aware sentence retrieval for nonspeaking individuals with motor disabilities.’ In Proceedings of the 38th ACM Conference on Human Factors in Computing Systems (CHI 2020). DOI: 10.1145/3313831.3376525