The importance of a standard based interoperability framework is now widely recognized. In order to understand the challenges, feasibility and impact of outcomes of such architectures for public health improvement, it is important to implement and demonstrate the usefulness of data analytics based on pathway provided by the framework. This work provides an implementation details and results of transferring dengue data from source database to data analytical programs used by public health authorities. Dengue fever, a mosquito-borne disease that occurs in tropical and subtropical areas of the world, is considered to be a significant threat in both developing and the developed countries. This study investigates the spatio-temporal distribution of dengue in Pakistan from the years 2014-2017 and identify the most frequently affected hotspots of dengue across the the 3 provinces in the country using data mining, clustering and GIS based analysis of data. In addition, the identification of most vulnerable dengue locations has led to investigation into the potential environmental or geographical conditions that may have contributed to dengue prevalence in the area. These investigations present interesting results that shows positive correlation for temperature, humidity and population density with the dengue incidences. In addition, the results are able to identify disease hotspots over time windows that can be mapped to any spatio-temporal scales. The outcome of this research is crucial for optimal use of resources for combating dengue fever at a regional or national scale by identifying the hotspots. The hotspots identified can be use to create a sentinel surveillance network for a pandemic disease. In addition, the work serves as a useful reference for a comprehensive national level early warning and rapid disease outbreak detection framework for any disease of public health context.
Read full abstract