中文摘要：

目的利用南京市2004-2009年呼吸系统疾病死亡病例和同期的气象资料,分析了气象因子（包括三种舒适度指数）与呼吸系统疾病死亡人数的相关性。方法通过BP神经网络建立了呼吸系统疾病死亡人数的预报模型,并对其进行评价。结果气象因素及其变化与呼吸系统疾病死亡人数有密切的关系。建立的呼吸系统疾病死亡人数的神经网络预报模型结构为15-20-1（即有15个输入、20个隐含节点和1个输出）,训练精度为0.005,训练了26步达到目的,预测准确率达80.11%。结论与统计预报方法相比较,该方法计算简便、误差较小、预测准确率高,对呼吸系统疾病死亡人数有较好的预测效果,为医疗气象预报提供了一种新方法,具有进一步的研究价值。

英文摘要：

Objective Using the data of respiratory system deaths and meteorological factors within the same time from2004 to 2009 in Nanjing,and analyzed the correlation between meteorological factors which include three human comfortable indexes and respiratory system deaths. Methods The back-propagation（ BP） artificial neutral network（ ANN） model was built and evaluated. Results The result showed： a close relationship exists between the meteorological factors and respiratory system deaths,the ANN predict model structure was 15-20-1,15 input notes,20 hidden notes and 1 output note. The training precision was 0. 005 and the final error was 0. 005 after 26 training steps. The results of forecast showed that predict accuracy over80. 11%. Conclusions Compared with statistical forecasting methods,this method is easy to be finished with smaller error,and higher ability on respiratory system deaths on independent prediction,which can provide a new method for medical meteorology forecast and have the value of further research.