Researchers at Khalifa University are developing mathematical tools that can analyze data generated by the numerous Covid-19 transmission models already in use.
In an effort to understand the dynamics of the Covid-19 outbreak, a team of researchers led by Dr. Dimitris Goussis, Professor of Mechanical Engineering at Khalifa University, is developing mathematical tools to support the analysis of numerous Covid-19 transmission models already in use. The findings are expected to support public health policies and efforts to tackle the virus’ spread and its impact on society.
Mathematical models try to forecast things such as how a disease spreads, the total number infected and the duration of the epidemic. They can estimate various epidemiological parameters. For example, they can predict the range of values that might vary the reproductive number – a term that indicates how contagious an infectious disease is. They can also assess the influence of different public health interventions in controlling the outcome of the epidemic.
“We’re planning on developing mathematical tools for identifying the degree to which various factors (like, social distancing or recovery rate) influence the development and control of the Covid-19 outbreak, and how they may prevent its resurgence,” explained Dr. Goussis. “This work will be based on existing and widely used simple compartmental mathematical models, which are frequently used in analyzing diseases spread by the transmission of viruses, bacteria or fungi.”
Dr. Goussis’ team will apply their analysis to existing disease models, including the compartmental models used to assess disease impact. Timescale analysis leads to deeper understanding by identifying the components of the model that drive the system and those that are unimportant. In this way, the team can assess which interventions are most beneficial to reducing disease spread, among other things.
Compartmental models for disease modelling began with Nobel Prize winning research on the transmission of malaria in African countries, early in the 20th century. They have also been used recently to assess the impact of vaccination against polio, Ebola and measles, and to study the spread of dengue fever, swine flu, norovirus and varicella-zoster virus. Compartmental models simplify infectious disease modeling by assigning the population to various compartments with different labels.
The simplest compartmental model is the SIR epidemiological model involving Susceptible, Infectious and Recovered labels. It is considered to be quite effective at predicting the spread of human-to-human infectious diseases. More detailed models involve more compartment labels, such as Exposed, Quarantined or Asymptomatic infected.
“A major portion of the Covid-19 literature is based on compartmental models,” explained Dr. Goussis. “Among the various features addressed are the international spread from Wuhan, the prediction of the reproduction number and the assessment of travel restrictions in reducing the number of infected cases, and the effect the virus has on different age groups. All of these works have addressed features of the problem, like profiles of the infected population, but have not examined the dynamics of the process and especially the issue of the timescales characterizing the various phases of the outbreak.”
“Given a model, timescale analysis reveals the mechanisms that control the evolution of the system,” said Dr. Goussis. “Our work will identify the components of the model under consideration that control the dynamics of the outbreak’s evolution and its possible resurgence; i.e. the contact rates, the infectious periods, and the impact of relaxing social distancing.”
19 August 2020