GIS Helps Model Infectious Diseases in Space and Time

GIS Helps Model Infectious Diseases in Space and Time

There is some good news in the most recent wave of Ebola outbreaks in Africa. The spread of the disease in Nigeria seems over. It was the first experience that Nigeria, the most populous nation on the continent, had experienced.

But as the overall experience has shown so far, staying ahead of a deadly disease is a difficult task. Transmission is a product of space and time. Public health experts have had to fight both geography and time to keep the spread from being far wider and deeper than it already has been.

That is why GIS training can be vital to epidemiology. The ability to track things by geography as well as time can become a vital tool. That is what makes some research at the University of North Carolina at Charlotte so interesting. A research team led by Dr. Eric Delmelle is investigating how to better model diseases in space and time.

Modeling is a way to mathematically replicate and approximate the actions of complex systems to better understand their current behavior and predict future activity. Such ability could be vital in anticipating the spread of disease and allowing officials to take action in time to curb outbreaks before they had a chance to run wild.

Delmelle works with a number of specialists in geography and GIS studies. Using such techniques as parallel computing and space-time visualization, the group is applying GIS and other theory to better understanding the spread of diseases. For example, one of its projects is “space-time visualization of dengue fever outbreaks”.

Dengue fever is a vector-borne infectious disease, which is known to quickly develop and spread under favorable weather conditions. This presence of the disease is also affected by societal behaviors (population migration and prevention attitude towards the disease). We apply a spatial and temporal extension of the kernel density estimation algorithm to map space-time clusters of dengue fever in Cali, Colombia for the year of 2010. We conduct the Space-Time Kernel Density Estimation (STKDE) in a parallel computational framework to account for the computational complexity involved. We extract intricate disease dynamics in an interactive manner, using a powerful 3D environment. Computational and cartographic impacts of varying space-time bandwidths are presented. The findings from our work can help better understand the dynamics of a quickly spreading disease.

The work is challenging, given the number of factors that combine to enable or modify the spread of diseases. But the payoff of success is literally a matter of life and death.