< Back to Article

Big Data Opportunities for Global Infectious Disease Surveillance

Figure 1

A schematic overview of the process of predicting spatial disease risk.

The definitive extent of infectious disease occurrence at the national level (red is certain presence, green is certain absence) [16] is combined with assemblies of known occurrence, presence points (red dots), to generate putative pseudo-absence points (blue dots). The presence and pseudo-absence data are then used in the analyses, with selected environmental covariates to predict disease risk, formally the probability of occurrence of the target disease. In this example a risk map of dengue is shown, shaded from low probability of occurrence in blue to high probability of occurrence in red [8]. The arrows represent data flows.

Figure 1