目的 了解2012—2023年合肥市急性出血性结膜炎（acute hemorrhagic conjunctivitis, AHC）流行特征并预测发病人数，为防控措施的制定提供科学参考。方法 从中国疾病预防控制信息系统收集2012—2023年合肥市AHC报告病例。对合肥市AHC的人群、地区和时间分布进行描述性分析；采用Prophet和季节性自回归移动平均（seasonal autoregressive integrated moving average, SARIMA）模型预测发病趋势。利用Jupyter Notebook软件建模分析，模型精度评估方法为均方根误差（root mean squared error, RMSE）、平均绝对百分比误差（mean absolute percentage error, MAPE）和平均绝对误差（mean absolute error, MAE）。结果 2012—2023年，合肥市共报告AHC病例2 837例，无死亡病例；年均发病率为2.88/10万，总体趋势为先降再升后平稳。发病年龄中位数为39（23，59）岁，发病性别以男性为主（1 570例，占55.34%），职业以农民（1 419例，占50.02%）为主，年均发病率前三位的县（市、区）分别是肥西县（1 545例，15.69/10万）、长丰县（338例，4.10/10万）和巢湖市（300例，3.19/10万），全年各月均可发病，发病高峰集中在5—9月。SARIMA模型中筛选出较优模型为SARIMA(1, 0, 1)(0, 0, 1)₁₂，Prophet模型（RMSE=10.42，MAPE=44.78%，MAE=7.68）的所有误差值均低于SARIMA(1, 0, 1)(0, 0, 1)₁₂模型（RMSE=17.68，MAPE=56.41%，MAE=15.04），具有更好预测效果。结论 2012—2023年合肥市AHC发病总体趋势为“下降-上升-平稳”，需加强对男性、农民及学生等重点人群的防控工作，尤其是在夏秋季和高发县区，需开展针对性监测和健康宣教。相较于SARIMA(1, 0, 1)(0, 0, 1)₁₂模型，Prophet模型在预测合肥市AHC发病数时效果更好，可作为短期预测模型。

Objective To understand the epidemiological characteristics of acute hemorrhagic conjunctivitis (AHC) in Hefei from 2012 to 2023 and predict the number of AHC cases, thus to provide scientific reference for the formulation of prevention and control measures. Methods The reported cases of AHC in Hefei from 2012 to 2023 were collected from the Chinese Information System for Disease Control and Prevention. Descriptive analysis was used to analyze the population, regional, and temporal distribution of AHC in Hefei. The Prophet model and seasonal autoregressive integrated moving average (SARIMA) model were constructed to predict the incidence trend of AHC. Modeling and analysis were performed by using Jupyter Notebook software. The model accuracy was evaluated by using root mean squared error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE). Results From 2012 to 2023, a total of 2 837 AHC cases were reported in Hefei, with no deaths. The average annual incidence rate was 2.88/100 000. The overall incidence trend decreased first， then increased and finally stabilized. The median age of onset was 39 (23, 59) years. The cases were mainly male (1 570 cases, 55.34%). Farmers (1 419 cases, 50.02%) constituted the primary affected occupations. The top three county-level divisions with the highest annual incidence rates were Feixi County (1 545 cases, 15.69/100 000), Changfeng County (338 cases, 4.10/100 000) and Chaohu (300 cases, 3.19/100 000). AHC cases occurred throughout the year in Hefei, but demonstrated a distinct seasonal peak between May and September. Among the SARIMA models, the optimal model selected was SARIMA(1, 0, 1)(0, 0, 1)₁₂. All error values of the Prophet model (RMSE=10.42, MAPE=44.78%, MAE=7.68) were lower than those of the SARIMA(1, 0, 1)(0, 0, 1)₁₂ model (RMSE=17.68, MAPE=56.41%, MAE=15.04), indicating better prediction effect. Conclusion From 2012 to 2023, the incidence of AHC in Hefei has shown a "decline-rise-stabilization" trend. It is necessary to strengthen the prevention and control efforts for key populations such as males, farmers and students, especially in summer and autumn and high-incidence counties and districts, targeted monitoring and health education should be carried out. Compared with the SARIMA(1, 0, 1)(0, 0, 1)₁₂ model, the Prophet model performs better in predicting the number of AHC cases in Hefei, and can be used as a short-term prediction model.