The new 4.2 version of the GLEAMviz Simulator has been released!

The 4.2 release allows the users to visualize on the geographical mapping the additional data layers that were introduced in version 4.1, regarding population and healthcare indicators. The 4.2 version also features major improvement to the simulation engine and enhancement to the data visualization settings.

The  GLEAMviz install package, the new manual and full documentation for the current version 4.2 can be downloaded here. If you already have any GLEAMviz 4.x version you can automatically update the client by launching the GLEAMviz application.

We are working on much more new features and improvements as well as new modeling capabilities. Thanks for using GLEAMviz!

Map visualization displaying the hospital beds data layer and the epidemic infection data layer for a single-run scenario.

A new release of the GLEAMviz Simulator client is available, version 4.1, which features some extra data visualizations.

• Additional demographic and health-related informations in the city info popups:
  • Average number of hospital beds per 1000 inhabitants
  • Average number of physicians per 1000 inhabitants
  • Average distribution of the population into age brackets
• Visualization of the cumulative individual transitions on the map as an alternative option to new transitions.

• Intuitive highlight of the census basins associated to each city/airport.

The new city popup displays additional health-related and demographic informations.

Some minor bugs have been fixed as well, providing a better usability both of the simulations builder and of the geographic mapping.
The updated version of the manual is available for download: GLEAMviz simulator client manual v4.1.

Real-time forecasting of unfolding epidemics represents an extraordinary challenge for mathematical and computational models due to the technical challenges posed by parameter estimation and validation. For this reason, the reliability of mathematical and computational models to support public-health decisions during an epidemic is currently under debate. The 2009 A/H1N1 pandemic represented an unprecedented opportunity to validate and assess real-time model predictions.

In the recently published paper:

Real-time numerical forecast of global epidemic spreading: case study of 2009 A/H1N1pdm
Michele Tizzoni, Paolo Bajardi, Chiara Poletto, José J. Ramasco, Duygu Balcan, Bruno Gonçalves, Nicola Perra, Vittoria Colizza, Alessandro Vespignani.
BMC Medicine, 10:165 (2012).

we present a thorough validation and assessment of the results obtained using GLEAM during the course of the 2009 A/H1N1 pandemic.

In detail, we used GLEAM in real-time to generate stochastic simulations of epidemic spread worldwide, yielding the incidence and seeding events at a daily resolution for 220 countries. Using a Monte Carlo Maximum Likelihood analysis, the model provided an estimate of the seasonal transmission potential during the early phase of the H1N1 pandemic, and generated ensemble forecasts for the activity peaks in the northern hemisphere in the fall/winter wave. The forecasts were published in September 2009, well before the peak weeks of epidemic activity in the northern hemisphere.

Now, we have validated those results against the real-life surveillance data collected in 48 countries, and their robustness assessed by focusing on 1) the peak timing of the pandemic; 2) the level of spatial resolution allowed by the model; and 3) the clinical attack rate and the effectiveness of the vaccine. In addition, we studied the effect of data incompleteness on the prediction reliability.

Real-time predictions of the peak timing are found to be in good agreement with the empirical data, showing strong robustness to data that may not be accessible in real time (such as pre-exposure immunity and adherence to vaccination campaigns), but that affect the predictions for the attack rates. The timing and spatial unfolding of the pandemic are critically sensitive to the level of mobility data integrated into the model. For a detailed comparison between the surveillance data on peak incidence and our forecasts, take a look at this page.

Left: peak weeks of the epidemic activity in the baseline stochastic forecast for seven Canadian Provinces, ordered eastward from top to bottom. Right: map of Canada, where the Canadian Provinces under study have been highlighted in red.

Our results show that large-scale models can be used to provide valuable real-time forecasts of influenza spreading, but they require high-performance computing. The quality of the forecast depends on the level of data integration, thus stressing the need for high-quality data in population-based models, and of progressive updates of validated available empirical knowledge to inform these models.

Difference of the median peak weeks in the reference stochastic forecast, where the full mobility dataset is considered, and the scenario that includes only the 500 busiest airports, as a function of the median peak week in the reference forecast. Dots are color coded according to the corresponding airport’s climate zone.

The GLEAMviz team is proud to announce a new release of the GLEAMviz tool, which features a complete rewriting of the server component and additional functionalities in the client application.
The Public Edition 4.0 of the GLEAMviz client is now available for download.

With this update we have introduced:

• New optimized engine code implementing the Global Epidemic and Mobility model. Your simulations will now run approximately ten times faster than before!
• Multiple choices for implementing local human mobility (data, gravity models, ABM, etc.)
• New redesigned interface for the simulations management that allows the setting of disease models and interventions that can be  geographically dependent (and for specific time period).
• New middleware component with a more structured application server and a standardized client-server communication protocol.

The 4.0 version of the GLEAMviz computational platform features also the integration with the Epidemic Marketplace data platform (http://www.epimarketplace.net), a data integration system for epidemiological resources. There you will find data and tools relevant for epidemic modeling uploaded by Marketplace user community.
The integration between the two platforms allows the users of the GLEAMviz client to upload and store their simulations on the Epidemic Marketplace. By means of the functionalities available on the Epidemic Marketplace website, it is also possible to share simulations with other GLEAMviz users.

We are pleased to announce the release of the next major version of the GLEAMviz Simulator. This version features a major update of the application’s interface, including the new Simulator Builder which integrates the former Simulation Wizard and Model Builder. The 3.0 version also provides two new visualisations: the SPaTo Visualization and the 3D Globe Visualization.

Example of the new SPaTo Geographic visualization

The EE² workshop: Epiwork/Epifor 2nd International Workshop – Facing the Challenge of Infectious Diseases took place in Courmayeur (Aosta, Italy) on January 18-20 2012. The event has been a great occasion to present the ongoing developments of the GLEAMviz simulator platform, showing the version currently available, discussing its future evolution, and presenting the scientific results reached within the project.

Alessandro Vespignani gave a talk titled Real time numerical forecast of global epidemic spreading using large-scale computational models: case study of 2009 A/H1N1pdm, illustrating the side by side comparison between the epidemiological surveillance data for the 2009 H1N1 pandemic peak of more than 40 countries in the Northern Hemisphere and the predictions obtained in real-time by GLEAM. The agreement of the model prediction and the surveillance data demonstrate the importance that computational tools such as GLEAM may have in facing the emergence of a pandemic event.

Corrado Gioannini illustrates the new features of the forthcoming GLEAMviz simulator 2.9 during the poster session

Furthermore, in his talk Alessandro presented the forthcoming 2.9 version of the simulator, including brand new visualization features and a redesigned user interface aimed at maximizing flexibility and usability. The participants of the workshop could also experiment live the new features of the forthcoming GLEAMviz 2.9 release, as Corrado Gioannini was providing a hands-on experience during the poster sessions. The exhibit gathered a lot of interest and many people stopped ad asked about the project.

Stay tuned for the 2.9 release!

The poster session of the workshop featured two posters based on the GLEAMviz project. The first poster, presented by Michele Tizzoni, focused on the study of the impact of travel restrictions in delaying and containing the spreading of an epidemic. The study, published in PLoS ONE, combines a theoretical analysis with a computational modeling work and, by considering the 2009 H1N1 pandemic as an example, shows the scarce effectiveness of such an intervention measures.
A second poster, titled Integrating the GLEAMviz Simulator tool with the Epidemic Marketplace platform, was presented by Corrado Gioannini and Joao Zamite, describing the effort carried out within the EPIWORK project aimed at providing a powerful but seamless integration of the GLEAMviz simulator tool with the Epidemic Marketplace information platform.

Human mobility networks, travel restrictions and the gobal spread of 2009 H1N1 pandemic

Integrating the GLEAMviz Simulator tool with the Epidemic Marketplace platform

GLEAMviz appeared today at the Top-IX Annual Conference Big Data in a Living Web in the keynote presentation by Alex Vespignani and in the invited talk by Marco Quaggiotto.

Can we foresee the start of a conflict? can we anticipate the spread of a pandemic pathogen? can we assess the systemic risk and the impact that a financial crisis, a nuclear disaster, or a pandemic event will have on our society and our planet?

GLEAMviz was presented by Alex as an example product of the Big Data Revolution, where high-resolution data on human behavior can be integrated into a disease model to anticipate the geographical and temporal spread of a pandemic influenza.

Marco presented the perspective of shaping the data produced by GLEAMviz into informative visualizations that help making sense of the simulated scenario and guiding the analysis of the epidemic pattern.

The Conference also saw invited talks by Alan Mislove from Northeastern University in Boston, USA, and Cesar Hidalgo from MIT in Boston, USA.
We thank Top-IX, torino piemonte internet exchange, for the invitation to the event.

We are pleased to announce the release of GLEaMviz Simulator v2.8. The principal new feature in this release is the ability to specify time-dependent overrides of the default values for variables defined in the compartmental model. Such overrides can be used to model time-dependent changes in the infection dynamics of the epidemic such as the effect of mitigation policies on the disease parameters.

New panel in the Simulation Wizard that enables the modeler to specify time-dependent variable overrides.

A new panel in the Simulation Wizard, shown in the figure below, enables the modeler to add one or more time-dependent variable overrides. Each override applies for one selected variable and is valid from the selected start date up to and including the selected end date. The overriding value is specified as an algebraic expression that may include references to other variables, just like a default value for variables or transition rates expressions.
Starting from this release it is now also possible to use parenthesis in such expressions.

A new public release of the GLEaMviz simulator is now available!
The main difference of this new update relates to the simulations’ output data that can be retrieved after execution to be visualized and analyzed. It is indeed now possible for GLEaMviz users to download from the server the simulation’s output of multiple compartments at the same time, separately, and then choose to visualize and analyze them one by one or aggregate them in a flexible way in the visualization window.

Detail of the Visualization Window of the GLEaMviz simulator showing the widget to choose the compartments.

Additional improvements have been added to this release, including the possibility to visualize an arbitrary number of plot charts in the visualization window and to immediately save the data of a plot chart to a file for subsequent analysis.

The new features have been documented in the updated version of the manual, which can be found here.
You are welcome to try our latest release and give us your feedback!