Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XL-2/W3, 19-23, 2014
https://doi.org/10.5194/isprsarchives-XL-2-W3-19-2014
© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.
 
21 Oct 2014
Risk Mapping of Cutaneous Leishmaniasis via a Fuzzy C Means-based Neuro-Fuzzy Inference System
P. Akhavan1, M. Karimi1, and P. Pahlavani2 1Faculty of Geodesy and Geomatics Eng, K.N Toosi University of Technology, No.13446, Mirdamad Cross, Valiasr st., Tehran, Iran
2Dept. of Surveying and Geomatics Eng., College of Eng., University of Tehran, Tehran, Iran
Keywords: Cutaneous Leishmaniasis, Data Mining, Fuzz C Means Clustering, Neuro-Fuzzy Systems Abstract. Finding pathogenic factors and how they are spread in the environment has become a global demand, recently. Cutaneous Leishmaniasis (CL) created by Leishmania is a special parasitic disease which can be passed on to human through phlebotomus of vector-born. Studies show that economic situation, cultural issues, as well as environmental and ecological conditions can affect the prevalence of this disease. In this study, Data Mining is utilized in order to predict CL prevalence rate and obtain a risk map. This case is based on effective environmental parameters on CL and a Neuro-Fuzzy system was also used. Learning capacity of Neuro-Fuzzy systems in neural network on one hand and reasoning power of fuzzy systems on the other, make it very efficient to use. In this research, in order to predict CL prevalence rate, an adaptive Neuro-fuzzy inference system with fuzzy inference structure of fuzzy C Means clustering was applied to determine the initial membership functions. Regarding to high incidence of CL in Ilam province, counties of Ilam, Mehran, and Dehloran have been examined and evaluated. The CL prevalence rate was predicted in 2012 by providing effective environmental map and topography properties including temperature, moisture, annual, rainfall, vegetation and elevation. Results indicate that the model precision with fuzzy C Means clustering structure rises acceptable RMSE values of both training and checking data and support our analyses. Using the proposed data mining technology, the pattern of disease spatial distribution and vulnerable areas become identifiable and the map can be used by experts and decision makers of public health as a useful tool in management and optimal decision-making.
Conference paper (PDF, 1938 KB)


Citation: Akhavan, P., Karimi, M., and Pahlavani, P.: Risk Mapping of Cutaneous Leishmaniasis via a Fuzzy C Means-based Neuro-Fuzzy Inference System, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XL-2/W3, 19-23, https://doi.org/10.5194/isprsarchives-XL-2-W3-19-2014, 2014.

BibTeX EndNote Reference Manager XML