Spatiotemporal Prediction of Crimean-Congo Hemorrhagic Fever Outbreaks in Ardabil Using Random Forest Model and Analysis of Climate Change Impacts (2015–2030)
Code: G-1875
Authors: Azam Nasrollahi *, Robab Hosseinpour ℗
Schedule: Not Scheduled!
Tag: Health Policy, Law & Management in AI
Download: Download Poster
Abstract:
Abstract
Background and Aims: Infectious diseases, including Crimean-Congo Hemorrhagic Fever (CCHF), pose significant challenges to global public health. Climate change may influence the dynamics of these diseases, necessitating accurate computational tools for outbreak prediction. This study aims to develop a spatiotemporal machine learning framework to predict CCHF incidence at the county level in Ardabil, Iran, from 2015 to 2030, while assessing the impact of climate change on outbreak patterns. Method: The proposed research will employ the Random Forest model to capture nonlinear relationships and temporal lags (1 to 12 weeks). Epidemiological data (2019–2023), meteorological data (temperature, relative humidity, precipitation), and environmental data (normalized difference vegetation index, land cover) will be collected from local and global sources. Climate change scenarios (RCP4.5 and RCP8.5) will be analyzed to forecast changes in high-risk areas. Results: It is anticipated that variables such as temperature and relative humidity will be identified as key drivers of CCHF outbreaks. Hotspot analysis is expected to highlight areas like Meshginshahr and Parsabad as high-risk regions. The model is likely to provide high predictive accuracy, enabling the development of an early warning system for targeted interventions. Conclusion: This study will offer a scalable and effective framework for predicting climate-driven zoonotic diseases, with potential applications for public health policy in Ardabil and similar regions.
Keywords
Crimean-Congo Hemorrhagic Fever, Random Forest,Climate Change
