A Geospatial Study On Malaria Risk Prediction And Mapping Of Integrated Tribal Development Agency Paderu Region India Using SAMRR

Abstract

quotGlobally risks of communicable and non-communicable diseases are increasing in recent newlinetimes due to climate change. For instance. Malaria and Covid-19 are best examples of newlinecommunicable diseases which are showing greater impact across the globe. According to newlinethe WHO 2022 report, approximately 274,000 children under the age of five died from newlinemalaria, accounting for 67% of all global malaria deaths. But United Nations (UN) emphasizes newlinethe third Sustainable Development Goal (SDG 3): Ensure healthy lives and promote well newlinebeing for all at all ages to be achieved by 2030. Under SDG 3, ending malaria by 2030 is of a newlinehigh priority challenge for the health administration. In the study area, tribal areas which are newlinenow placed in Alluri Seetharamaraju and Anakapalli districts are very prone to malaria. Even newlinethough the Study Area Annual Parasite Indices (SAPI) impact was reduced gradually from newline6.87777 to 1.755217 during 2016-2020 period, but the API of study area is still 88.2% higher newlinethan Country Annual Parasite Index of 3.215849. The SAMRR is a geospatial predictive newlineanalytics machine learning tool that considers environmental factors such as temperature newline(T), rainfall (Rf), normalized vegetation index (NDVI), and water bodies (WB), and finds a newlinecorrelational analysis between these environmental factors and malaria cases in the study newlinearea, which is segregated into 12 mandals and 37 Primary Health Centers. Based on the high newlinesignificance of the correlational analysis between environmental factors and malaria cases, a newlinenovel SAMRR predictive factor was proposed with a designated range of values for each newlineparameter. With the SAMRR predictive analysis out of 12 mandals, it was found that 10 newlinemandals are at high risk and 2 mandals are at low risk, and the classification accuracy is 100% newlineand the AUC is 1.0. The classification performance of SAMRR is compared with the existing newlinemalaria prediction techniques like SARIMA (94 %), Maxent (90 %), Bayesian Decision newlineNetworks (79.1 %), and Generalized Linear Models (63.6 %). Am