Epidemiological Modeling

Epidemiological models are compartmental models that simulate the evolution of infectious disease outbreaks. The population is divided into compartments, e.g. Susceptible, Infected and Recovered (SIR), and the movements of subjects between compartments are defined by means of ordinary differential equations. The results of an epidemic simulation strictly depend on the model’s parameters; some of these parameters are biological and bound to the infectious disease, others are related to demographics and specific areas. As an example, the density, mobility and the social distancing measures specific to a country significantly affect the transmissibility of the infection. Besides estimating the SIR parameters, the traditional compartments can be extended to further model more complex dynamics.

This lecture will introduce the simplest compartmental models. It will also show how it is possible to extend the traditional SIR model, with the final aim of simulating outbreak scenarios under different policies and country’s specifics. For example, by modeling the mobility between infected areas it is possible to simulate the closure or the re-opening of specific zones, and the relative change in the overall infections rate. Additional compartments can also be added to estimate the hospitalization volumes, or the ratio between reported and unreported cases. Finally, having time-varying transmissibility allows for modeling the effect of lockdown measures. A complementary aspect of SIR extensions is the parameter fitting; complex models usually involve an increased amount of parameters, while only a few variables of the system are observable, such as the new cases identified daily.

Speakers:

• Dr. Stefano Giovanni Rizzo

• Dr. Mohamad Saad

• Dr. Sanjay Chawla

Speaker Biographies:

• Stefano Giovanni Rizzo

  Has been a Postdoctoral Researcher at Qatar Computing Research Institute (QCRI) at HBKU since July 2018 working on Deep Learning research problems in the areas of predictive analytics, urban mobility and anomaly detection. He has worked on Machine Learning research projects since obtaining his Master’s degree, in the Computer Science Department of the University of Bologna, Italy.

• Mohamad Saad

  Is a Research Scientist at QCRI. He joined in February 2017 and works on topics in statistical genetics, biostatistics, and bioinformatics. He obtained his PhD in Statistical Genetics/Bioinformatics from the University of Paul Sabatier III, Toulouse, France. One of his main research interests are combining different types of genomic data to explain complex diseases and the move towards precision medicine.

• Sanjay Chawla

  Is the Research Director of the Data Analytics Group and the Qatar Center for Artificial Intelligence (QCAI), QCRI. His research interests are data mining, machine learning and spatial data management.

If you've enjoyed this content, click below to find out more about the Publisher: Qatar Computing Research Institute (QCRI)

• Website:

  Qatar Computing Research Institute (QCRI)'s Website

• Social Media Accounts:

  Twitter