CLAIRE COVID19 topic groups

Leading AI researchers from CLAIRE are collaborating across borders to help address the COVID-19 pandemic

Epidemiological data analysis

This research group works on different types of models for epidemics, ranging from high level compartment models to agent based models, and how they can be used to study the dynamical aspects to improve complex decision taking on the effectiveness of prevention strategies. On the one hand, this involves model fitting and optimisation, on the other hand, learning and optimisation of prevention strategies, using epidemiological models as simulation environments.

Work is underway to identify collaboration mechanisms and structures, considering the support AI can offer in decision-making. This recognises the multi-criteria nature of the problem, balancing the needs of different stakeholders all of whom should be involved. 


Ann Nowé, AI Lab, Vrije Universiteit Brussel, Belgium. View details

Mobility and monitoring data analysis

This work sets out to understand the symptoms progression through self-reported data and its integration with mobility to forecast healthcare decision making. The goal is the development of an AI multilayer learning approach capable of creating evidence based knowledge, using complex networks for self-supervised learning, spatial temporal analysis and deep learning.
Work is underway to understand the data collected under the several self-reporting systems and test how useful the self-reported data is to forecast events. Initial models have been produced using different methodologies and compared with the officially reported statistics.


Jose Sousa, Faculty of Medicine, Health and Life Sciences, Queen’s University Belfast. View details

Bioinformatics (protein and molecular data analysis)

The topic joins forces from AI, clinical and life-sciences experts working on the analysis of complex and multi-sourced biomedical data integrating clinical evidence on COVID-19 with genomic and proteomic information, as well as molecular data. We are exploring data-driven AI methodologies and bioinformatics approaches covering network data analysis, machine learning and deep learning for graphs, predictive modelling, and feature selection of Omics data. Our primary goal is to support the community with the release of resources for

  • characterizing the disease from its related structural information, including prediction of viral protein folding
  • studying interactions between the virus and human hosts, including analysing protein-protein interaction data
  • filtering, retrieval, and generation of targeted drugs, leveraging molecular and well as proteomic information
  • delivering predictive insights onto the genetic features of the virus.

To enable these objectives, we are assembling a resource that fuses information from heterogeneous sources and different studies from the literature into a unique network-based representation, facilitating the use of relational and graph-based learning methods. We urge volunteers within all the relevant fields of AI and, even more importantly, domain experts, including, clinical specialists, immunologists, virologists, biologists, geneticists cheminformatics experts.


Davide Bacciu, Computational Intelligence and Machine Learning Group, Universita’ di Pisa, Italy. View details

Image analysis (CT scans, X-ray)

The group has two main goals: 1) to distil the current state of the art of methodologies and data sets for AI-assisted diagnosis of Covid-19 by way of imaging (e.g. TC Scan and X-ray) aiming at making diagnosis faster, cheaper and more manageable in the hospital processes (e.g. using low-resolution images). 2) to contribute to improving multidisciplinary knowledge by cross-breeding knowledge in computer science and radiology aiming at creating better, more informative reference datasets, together with data-gathering strategies, beyond the current outbreak.

The group builds on top of currently active EU-funded projects, such as 15M€ DeepHealth and 4.5M€ HPC4AI.


Marco Aldinucci, Computer Science Dept, University of Torino, Italy. View details

Social dynamics and networks monitoring

This work uses AI models to analyse social media data together with social, behavioral and economic data for two main purposes:

  1. Monitor social dynamics to analyse the COVID-19 “infodemic,”  defined as “an over-abundance of information – some accurate and some not – that makes it hard for people to find trustworthy sources and reliable guidance when they need it,” with the goal of identifying, monitoring and analysing the overload of unreliable information; of collaborating with data providers to obtain free access to relevant data; and of creating an interdisciplinary hub of experts to fight the “infodemic”; and
  2. Develop early-warning signals to support policy, informed by spatio-temporal analysis of emotions and sentiments; quantifying and modelling the socio-behavioural response.

Social media are playing a crucial role for spreading information, both reliable and unreliable, during the COVID-19 pandemic. Efforts are devoted to unravel the role played by both humans and software-assisted (i.e., social bots) in disseminating false or inflammatory content for social manipulation, a phenomenon recently discovered during political events, with the ultimate goal of attracting or driving collective attention towards a specific information.
Products of the individual team members, such as the infodemic observatory model developed by the topic coordinator within the Complex Multilayer Networks Lab at FBK, allow to monitor the current infodemic globally, in each country, or at sub-regional resolution in real time. Information, complemented with the analysis of cognitive content, based on natural language processing and computational psycholinguistics, might help to shed light on mass psychology and socio-behavioral response to the pandemic. Results can be used to support policy and decision makers with adequate and zone-specific actions.

Such tools can be disseminated and further developed with the support of the entire research team.


Manlio De Domenico, Head of Complex Multilayer Networks Lab FBK – Fondazione Bruno Kessler. View details


Work in this area investigates possible uses of robotic systems and robotic technologies in response to the current COVID-19 emergency and to its aftermaths, as well as strategies to improve technological preparedness to possible future crises Specifically, this team has studied: the use of mobile robots for disinfection of environments; specialized laboratory robots for biological tests and drug development; telepresence robots for social and medical assistance; manufacturing robots for flexible production. These uses of robotic technologies are in line with a recent editorial in Science Robotics. 

The group maintains a catalogue of robotic offers and demands relevant to the COVID-19 emergency, and it is liasing active research laboratories across Europe. We have found that the liaison aspect is especially important during a crisis, when access to laboratory resources and material may be seriously limited. It is also supporting euRobotics (the association of European robotic stakeholders) in writing a white paper on the potential usage of robotic technology in the COVID-19 emergency.


Alessandro Saffiotti, AASS Cognitive Robotic Systems Lab, School of Science and Technology, Orebro University, Sweden. View details


Scheduling and resource management

The group working on this topic has focused on automated planning and scheduling, and resource management in healthcare systems leveraging AI (deductive) methodologies and tools. An initial assessment of relevant resources has been completed, and a review of relevant publications, data and projects is underway. In addition, collaboration with the Galliera hospital in Genova, Italy, is underway to assist with workforce scheduling and automated planning of the utilisation of operating rooms with scarce resources and equipment.


Marco Maratea, Dipartimento di Informatica, Bioingegneria, Robotica e Ingegneria dei Sistemi. University Genova, Italy. View details