摘要
目的 分析淮河流域疟疾流行状况与MODIS卫星遥感图像中标准化植被指数（NDVI）的关系，探索MODIS卫星遥感图像在大范围疟疾疫情监测中应用的可能性。方法收集淮河流域70个县市的疟疾疫情资料；利用ArcGIS 8．3软件提取各县市的MODIS NDVI值，并制作疟疾发病率及NDVI专题图；以相关分析及logistic回归分析研究疟疾流行状况与NDVI值的关系。结果 2005年淮河流域疟疾发病率为2．50／万，8-10月为发病高峰，男女病例之比为1．49：1。相关分析显示：年均NDVI、年均极大NDVI，以及月均NDVI超过130、135、140的月数等变量与年发病率呈正相关（P〈0．05）。当月的月均NDVI，上个月的月极小NDVI、月均NDVI及月极大NDVI（OR=1．151、0．977、0．923、1．142，P〈0．01）等4个变量进入用于判断某县市是否进入流行月份的logistic方程。结论 淮河流域疟疾发病状况与NDVI值存在较明显的正相关，可利用MODIS遥感图像通过提取NDVI值来监测该流域疟疾的流行情况。
Objective To analyze the relationship between NDVI （normalized differential vegetation index） and malaria endemic in Huaihe River Valley, and explore the possibility of application of MODIS remote sensing images in large-scale malaria incidence surveillance. Methods Malaria endemic data of 70 sample counties in Huaihe River Valley were collected. ArcGIS 8.3 software was used to extract NDVI from MODIS satellite images. Correlation analysis and logistic regression analysis were used to study the relationship between NDVI and malaria endemic. Results The malaria prevalence in 2005 was 2.5/10 000 with a peak in months from 8 to 10, and the male to female ratio was 1.49 to 1. Correlation analysis indicated that the vegetation indexes including the annual mean NDVI, annual maximum mean NDVI, and the number of months of which the monthly mean NDVI was greater than 130,135 and 140 correlated with the malaria annual incidence （P 〈 0.05 ）. In order to explore the correlation of prevalent months to NDVI, a logistic regression function was established, including the monthly mean NDVI, the minimum, mean and maximum NDVI of last month with OR equals to 1. 151,0.977,0.923 and 1. 142 respectively （P 〈 0.01 ）. Conclusion Malaria endemic showed positive correlation with MODIS NDVI value, suggesting that MODIS images can be used for monitoring large- scale malaria endemic in Huaihe River Valley.
出处
《国际医学寄生虫病杂志》
CAS
2007年第3期122-124,共3页
International JOurnal of Medical Parasitic Diseases
关键词
疟疾
流行病学
标化植被指数
Malaria
Epidemiology
Normalized Differential Vegetation Index
