2013 Conference of the International Medical Geology Association (25–29 August 2013)

Paper No. 3
Presentation Time: 11:40 AM


SARFRAZ, Muhammad Shahzad, Computer Science, FAST-National University of Computer & Emerging Sciences, Chiniot-Faisalabad Campus, Chiniot, 35400, Pakistan, TRIPATHI, Nitin Kumar, Remote Sensing & GIS, Asian Institute of Technology (AIT), P.O. Box: 4, Klong Luang, Pathumthani, 12120, Thailand, KITAMOTO, Asanobu, Digital Content and Media Sciences Research Division, National Institute of Informatics, 2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo, 101-8430, Japan and SOURIS, Marc, Spatial Analysis for Epidemiology, Institut de Recherche pour le Développement (IRD), UMR 190, Marseille, 13055, France, shahzad.sarfraz@nu.edu.pk

In recent years, human community is facing challenge of health-related issues, and increase in infection diseases and their outbreaks are witnessed. Increased frequency and intensity of such outbreaks are causing massive loss of human life. Favourable environment, being the main factor, is playing a vital role for any type of disease outbreaks. Dengue fever is one of such diseases whose spread depends mainly on availability of favourable breeding sites. Conventional methods including surveys for observing and studying breeding places is very complex and time-consuming. On the basis of dengue indices, i.e. container index (C.I.), house index (H.I.) and Breteau index (B.I.) approaches collected three times a year during 2009 to 2011 for Ministry of Public Health department Phitsanulok, Thailand. These indices were adopted to ascertain the factors influencing dengue breeding habitats, so that these can be mapped from space. The most probable factors were temperature, humidity, rainfall, population density, elevation and land-use/land cover. All types of parameters were derived from freely available satellite images. Parameters synchronization and predication algorithm were developed on the basis of data mining decision tree method with freely available software and fuzzy logic approach. This model has the capability to improve its accuracy by adding more sampling data or expert opinion to predict favourable hotspots. Final algorithm can be useful to predict near real -time scenario in dengue habitats estimation. Furthermore, this study can be used to develop a monitoring mechanism for other type of diseases as well which are influenced by environmental and climatic factors, which is a potential and quick method to identify outbreak hotspots for early warning systems. Freely available remotely sensed images are potential sources to cope with such diseases in developing countries, and a big contribution of Earth Observation Satellites (EOS) that will open a new trend for sustainable society.