As the COVID-19 pandemic continues to envelop the world, Roberto Desimone considers how emerging technologies can help resolve both the current crisis, as well as prevent future outbreaks from taking root
How can I help?
Such has been the scale of the global COVID-19 pandemic that we have all had cause to pause, take stock and consider how we can each play our part in resolving the crisis. From social distancing to volunteering, supporting the most vulnerable to donating to charities, there are myriad ways that we, as individuals, can help flatten and then crush the curve.
But of course it’s not just about humans. Time is of the essence to solve this pandemic from both a human and economic standpoint and we must deploy the latest technologies to shorten the length of it.
Emerging technologies, such as Artificial Intelligence (AI), machine learning (ML) and quantum have a lot to offer to tackle the current and future pandemics. AI and ML should be exploited now to support better diagnosis and testing; to analyse complex data to find epidemiological patterns; and to support better coordination of crisis management, especially distribution logistics and resource management.
In the future, more sophisticated AI and ML (exploiting semantic graphs) and emerging quantum computing techniques will be able to address more complex simulation and modelling of the progression of pandemics through global communities, as well as consequences of global crises on business/finance, infrastructure/supply chains and government.
Here are five ways these current/emerging technologies can help to tackle coronavirus and other pandemics/global threats.
1. Better clinical diagnosis/testing regimes
Existing ML techniques, such as convolutional neural networks, are able to isolate key features in massive datasets to help clinicians provide better diagnosis through enhanced imagery analysis. They also improve testing by helping them discriminate key factors through concepts learned by various ML methods. By exploring massive datasets of images and other clinical data, these ML techniques can pinpoint critical features and factors often missed by novice practitioners.
2. Improved analysis of epidemiologic patterns
Deep learning, especially recurrent neural networks and emerging semantic learning techniques, such as explanation-based learning, can support complex (time-series) pattern analyses. These can be used to hypothesise epidemiologic (behavioural) patterns that explain how epidemics get transmitted across populations. Recent hierarchical task network (HTN) plan recognition approaches are able to manage thousands of competing hypotheses to provide better explanations of the current health threat landscape for individual viruses.
3. Better coordination of health crisis management
HTN plan generation methods have already been applied by defence to support military crisis management tasks, and related AI planning/scheduling techniques are being explored by different industry sectors to support improved distribution logistics and resource management. These approaches go beyond current project management systems and could be employed to support better resource allocation within hospitals and the health sector (such as ambulances/pharmacies/care workers); improved distribution of critical resources across health networks to plug gaps and better prediction of resource needs across NHS networks.
4. Improved risk management/contingency planning for future global pandemics/threats
Further exploitation of AI/ML techniques could better characterise key risks and how they relate to each, not only at the tactical-level, but also operationally and strategically. They can highlight how risks propagate and escalate across domains ranging from health to transport to finance, as well as how they affect complex human interactions and behaviours. Emerging quantum computing techniques could also address problems of greater orders of magnitude than today’s conventional computers and promise the ability to optimise not only the recognition of global epidemiological threats, but also mitigations that reduce threats across multiple sectors.
5. Radically enhanced simulation/modelling of progression of threats through global communities
Existing AI techniques for generating and managing distributed interaction simulations have already been demonstrated within military synthetic environments, supporting training, mission rehearsal and war-gaming. These could be applied to support the modelling and simulation of diverse health threats and pandemics. In the future, more sophisticated simulation and modelling methods exploiting quantum simulation and computing algorithms could enable more radically-enhanced simulation and modelling, not only of individual health pandemics, but also how these affect wider financial, business, transport and criminal activities globally.
At this time of global tumult, where bad news abounds and stories of human tragedy dominate the discourse, it’s important to also acknowledge that some signs of hope are slowly materialising.
Of course, we have a huge way to go before we have fully addressed the COVID-19 pandemic but a combination of human and technological ingenuity and purpose offers the optimal route back to normality.
Explore Government Insights
Stay up to date with the latest thinking, trends, technologies and projects from our Government teamFind out more
About the author
Professor Roberto Desimone is Manager, Strategic Innovation (Disruptive Technologies) at BAE Systems Applied Intelligence. He is also funded by the Royal Society as an Entrepreneur in Residence at the University of Bristol for quantum engineering, and visiting professor at Loughborough University.
- Transformation in the Time of Coronavirus. The Coronavirus has turned our lives upside down but that’s not all, says Mivy James. It’s also highlighted the plight of the digitally excluded, as well the systemic changes which should be made permanent, not temporary
- COVID-19 and the implications for fraud, credit risk and money laundering. How soon will we understand the COVID-19 impact on fraud, credit risk and money laundering? Mariola Marzouk investigates the steps banks can take in the current environment
- From bias to balance in Artificial Intelligence. Machine learning has the potential to reshape our world but biased algorithms cast a long shadow, says Mivy James. She explains why only algorithms free of human prejudices will unlock the full benefits of this new technology
- Making the most of machine learning. How can machine learning strengthen the use of data and alert investigation? We’ve got some ideas but we want to hear what *you* think too. Matt Boyd explains