Spatiotemporal Analysis and Data Mining of the 2014–2016 Ebola Virus Disease Outbreak in West Africa

This study investigates the spatiotemporal pattern of the 2014 Ebola virus disease (EVD) epidemic in the most heavily affected countries in West Africa and also mines the spatial associations between such pattern and other geographically distributed factors. Utilizing the publicly available open-source data, this study demonstrates a research design that integrates various geospatial data processing, analysis, and data-mining techniques to achieve the research objectives. For the 2014 EVD epidemic, spatiotemporal patterns were analyzed and visualized. Fine-grained population data were obtained through a population interpolation method to conduct healthcare accessibility analysis. Finally, associations between the spatiotemporal patterns of the incidences and healthcare accessibility as well as other factors were examined. The results suggest that (1) poor accessibility to healthcare facilities and EVD clusters are identified in many urban areas as well as some remote areas; and (2) EVD cases were more likely to be found in border areas of these countries where accessibility to healthcare facilities is poorer.