Spatiotemporal trends of foot and mouth disease (FMD) in Bangladesh from 2017 to 2023 and their associations with climatic factors and machine learning (ML) based prediction.

Foot-and-mouth disease (FMD) is a highly contagious transboundary viral disease that affects cloven-hoofed livestock, causing substantial economic losses. It is caused by an Aphthovirus under the family Picornaviridae. This study aimed to analyze the spatiotemporal distribution of FMD in Bangladesh, evaluate the influence of meteorological factors, and forecast future disease trends. FMD cases and meteorological variables were obtained from DLS and BMD, respectively. Spatial autocorrelation metrics, including Getis-Ord Gi*, Moran's I, LISA, and inverse distance weighting, were applied to identify disease clusters and risk zones. Correlation and regression analyses identified significant associations between FMD incidence and climatic factors. Relative humidity and temperature exhibited positive correlations with disease occurrence. Additionally, regression modeling revealed that both relative humidity and wind speed had a significant impact on FMD incidence. Predictive models, including ARIMA, Random Forest, and XGBoost, were applied, with XGBoost providing the most accurate forecasts (RMSE = 153.64), followed by Random Forest and ARIMA. FMD incidence peaked in March, with persistent southeastern hotspots and emerging northern clusters. IDW mapping showed elevated risk in southern regions, particularly during the pre-monsoon period, influenced by climatic factors. These insights can inform targeted surveillance and control strategies to mitigate the burden of FMD in Bangladesh.