IEEE Computational Intelligence Magazine - November 2020 - 52

of advanced data analytics techniques and
visual analytics tools to support policy decision-making in public health institutes across
Europe [7]. Further details, code repositories
and demonstrators are openly available at
www.midasproject.eu.
The MIDAS intelligent system unleashes
the potential of a range of analytics to explore
the confidential and sensitive data owned by health authorities
within a safe and secure environment. The system achieves
this through data visualization widgets, encapsulating the
analytics processes, within user-customizable dashboards that
summarize each user's selected output priorities, such as monitoring mental health or child obesity across the population
within the COVID-19 confinement restrictions. The main
contribution of this paper is the description of the MIDAS
platform resources, how these were rapidly refocused to address
COVID-19-related public health priorities, and how they can
help researchers and public health policy-makers achieve a
deeper and more complete understanding of the SARSCOV-2 disease.
The MIDAS integrated platform consists of software developed for data ingestion, processing, analytics and visualization.
These services are deployed locally within health authority
sites. The platform offers a bespoke co-created collection of
software and services enabling secure access to and exploration
of sensitive health data as well as open and social data brought
in from external applications, via the MIDAS dashboard. A
common user authentication service allows single sign-on
across the services. Data downloads are prohibited to mitigate
privacy concerns and risk. If a MIDAS user, such as a policymaker, requires access to the raw and/or prepared data, they
must make a data access request to the data gatekeepers to

The MIDAS intelligent system unleashes the potential
of a range of analytics to explore the confidential and
sensitive data owned by health authorities within a
safe and secure environment.
and research institutions have rapidly refocused their monitoring platforms to assess the impact of this common threat.
Examples include the WHO COVID-19 Dashboard by ArcGIS [2], the Coronaviruswatch platform by the UNESCO
Research Centre for Artificial Intelligence (IRCAI) [34] and
RavenPack's Coronavirus News Monitor [29], alongside
many other global and local initiatives. Such platforms enable
patterns and trends in disease behavior monitored throughout the population. They can also inform public health measures to reduce transmission and allow the impact of these to
be assessed.
The novel MIDAS public health platform [21], presented
in this paper and shown in Figure 1, goes a step beyond existing platforms, particularly in responding to the coronavirus
pandemic, by providing its users in public health authorities
with insightful information from a combination of sources
including world news, social media and published science,
alongside local public health data from the health institution
itself and other relevant data sources. The MIDAS platform
was co-created with academia, industry, and crucially, health
professionals, policy-makers, public health authorities and citizens, to align innovative technology with concrete public
health priorities and workflows [4]. It was developed to connect typically heterogeneous, isolated health data, and integrate
it with additional social data sources, to enable the application

FIGURE 1 The MIDAS platform dashboard refocused to explore COVID-19 over proprietary data, worldwide news, social media campaigns and
biomedical research.

52

IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | NOVEMBER 2020
http://www.midasproject.eu
IEEE Computational Intelligence Magazine - November 2020

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - November 2020
