Prediction of brucellosis incidence in China’s five highest-incidence provinces: Comparing time-series models with multi-source environmental predictors
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Brucellosis is a severe zoonotic disease with pronounced seasonality and regional heterogeneity in high-incidence areas of China. Reliable forecasting tools are needed to inform prevention strategies, but the optimal modeling approach across different regions remains unclear.
Principal Findings
We collected monthly brucellosis incidence and 17 environmental variables from 2014 to 2024 across five high-incidence provinces: Inner Mongolia, Xinjiang, Shanxi, Heilongjiang, and Hebei. A three-step procedure—cross-correlation analysis, multicollinearity diagnostics, and stepwise regression—was used to select exogenous predictors. We then compared four time-series models: seasonal autoregressive integrated moving average (SARIMA), SARIMA with exogenous variables (SARIMAX), long short-term memory (LSTM), and LSTM with exogenous variables (LSTMX). All five provinces showed a unimodal seasonal pattern with peaks between April and July, though environmental drivers and optimal lag periods varied substantially by region, ranging from 1 to 6 months. In forecasting performance, LSTM achieved the highest accuracy in Shanxi (R²=0.925), Hebei (R²=0.876), and Xinjiang (R²=0.829), outperforming SARIMA and SARIMAX. LSTMX performed best in Inner Mongolia (R²=0.759) and Heilongjiang (R²=0.772) but showed weaker performance than LSTM in Shanxi and Hebei. Overall, adding exogenous variables did not consistently improve predictions across provinces.
Conclusions
Our findings demonstrate that LSTM-based models offer clear advantages for brucellosis forecasting in most high-incidence provinces, but the value of incorporating environmental predictors is region-dependent. These results support the development of tailored early warning systems and precision prevention strategies for brucellosis in high-risk areas of China.
Author Summary
Brucellosis is a bacterial disease that spreads from animals to humans, posing serious health risks, especially to livestock workers. Although preventable, its incidence has been rising in parts of China, highlighting the urgent need for more accurate forecasting tools. Traditional statistical models often fail to capture the complex patterns driving disease outbreaks, so we turned to deep learning. We developed an LSTM model—a form of artificial intelligence that learns from historical data—and compared its performance against standard models across five high-incidence provinces. Our LSTM model performed remarkably well, explaining up to 92.5% of the variation in brucellosis cases in some high-risk areas. Interestingly, adding environmental factors like temperature and humidity improved predictions only in certain provinces, not all. This finding underscores that no single model works everywhere; the best approach depends on local conditions. Our work advances the development of smarter, real-time early warning systems for brucellosis and other zoonotic diseases, aligning with the One Health approach that recognizes the interconnectedness of human, animal, and environmental health.
