Objective To explore the application of the long short-term memory (LSTM) model and the autoregressive integrated moving average (ARIMA) model in predicting varicella incidence trends, thereby providing data support for targeted early warning measures, evaluation of vaccination strategies, and prevention and control efforts.

Methods Weekly incidence data of varicella from January 2014 to March 2024 in Qingyang were selected to construct both LSTM and ARIMA models. The constructed models were utilized to fit and compare the incidence numbers from March 2024 to March 2025. Model performance was evaluated using root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and the coefficient of determination (R²). Finally, the optimal model was applied to predict weekly varicella incidence from 2025 to 2026.

Results A total of 10 075 varicella cases were reported between 2014-2024, with the reported incidence rate increasing from 19.74/100 000 in 2014 to 93.18/100 000 in 2024, resulting in an average annual reported incidence rate of 41.27/100 000. The incidence rate was significantly higher in males (44.20/100 000) than in females (38.43/100 000) (χ²=49.070, P<0.001). Cases were predominantly concentrated in individuals ≤18 years old, accounting for 85.94% (8 658 cases) of the total. students represented the primary affected group with a total of 5 504 cases or about 54.63% of all reports. The R²value for predictions made by the LSTM model on the test set was found to be 0.92, with RMSE at 6.22, MAE at 4.76, and MAPE at only 0.20. Conversely, for ARIMA(2,1,1)(2,0 ,2)52 on its test set yielded an R² value of -0.03, RMSE at 26.29, MAE at 18.72, and MAPE at 0.50. The results indicate that the LSTM model outperformed the ARIMA model across all four evaluation metrics (R², MAPE, RMSE, and MAE), demonstrating its superior predictive accuracy.

Conclusion The LSTM model demonstrates superior predictive performance compared to ARIMA models, providing a theoretical foundation and practical guidance for early warning measures, vaccination immunization, and control efforts related to varicella infections.