Epidemiological Characteristics And Prediction Of Measles Incidence Trend In Urumqi

Objective:We analyzed the epidemiological characteristics of measles cases in time regions and populations in Urumqi from 2011-2021,explored the application of seasonal Autoregressive Integrated Moving Average model and BP neural network in the prediction of measles and incidence in Urumqi,and verified the feasibility and applicability of the model,compared the prediction accuracy of the model,and derived the optimal model to provide an important epidemiological basis for the prevention and control of measles in Urumqi.Methods:In this study,data were processed using Excel,and the three-interval distribution of measles in Urumqi city from 2011-2021 was analyzed using descriptive epidemiological methods,and month-by-month statistics of measles incidence in Urumqi city from 2011-2021 were performed using SPSS 21.0 forX²test,trend test R software was used to build ARIMA and BP neural network models,and the predictions of measles incidence in Urumqi for 2022-2026 predicted by the two models were compared by comparing Mean Absolute Error（MAE）and Root Mean Square Error（RMSE）Error（MAE）and Root Mean Square Error（RMSE）were used to evaluate the prediction effect of the models.Results:A total of 1271 measles cases were reported in Urumqi from2011-2021,with an average annual incidence of 3.68/100,000,and the incidence of measles in Urumqi has been decreasing year by year.Diaspora children,domestic and non-working,children aged 0 to 2 years and people aged≥19 years were the high incidence of measles.The optimal model was established as ARIMA（2,1,1）（2,0,0）₁₂All parameters of the model were significantly different（P<0.05）,and the actual values were within the 95%confidence interval of the predicted values.The monthly incidence rates of measles in Urumqi city for 2022-2026 were predicted with MAE and RMSE values of0.130 and 0.205,respectively.5-10-1 was determined for the BP neural network model as the final model,and the predicted trend of the model was basically fitted to the actual values.The monthly incidence rates of measles in Urumqi for 2022-2026 were predicted with MAE and RMSE values of 0.015 and 0.006,respectively.Conclusion:The ARIMA model and BP neural network model were used to predict the incidence of measles in Urumqi from 2022-2026,and the results both showed that the incidence of measles in Urumqi from 2022-2026 showed a fluctuating upward trend year by year,with the BP neural network model showing a larger increase in 2025 A significant upward trend was observed.The incidence rate was the highest in 2026,at 0.0697/100,000.