From Epidemic Modeling to Forecasting: Understanding Cholera Outbreaks in Malawi

Cholera continues to pose a serious risk in developing areas, necessitating strong forecasting to guide public health measures. This research merges qualitative dynamics with machine learning to estimate cholera transmission patterns in Malawi. An epidemic model based on mechanistic principles captures the spread of the disease through parametric calibration. Sensitivity analysis using partial rank correlation coefficients pinpoints critical parameters that affect transmission. The basic reproduction number defines the long-term trends, while bifurcation analysis illustrates how disinfection influences the stability of the disease. To improve predictive accuracy, we combine the mechanistic model with ARIMA and autoregressive neural networks, creating hybrid machine learning models informed by the epidemic context. We generate short-term predictions of cholera cases, showing the advantages of integrating temporal disease dynamics into data-driven approaches. This combined method provides a replicable framework for making forecasts and aids in timely decision-making for epidemic management.

From Epidemic Modeling to Forecasting: Understanding Cholera Outbreaks in Malawi

Monday, July 14 at 6:00pm

Parthasakha Das

"From Epidemic Modeling to Forecasting: Understanding Cholera Outbreaks in Malawi"

#SMB2025 Follow