Incidence Rate of Hypertension Forecasting Based on Generalized Regression Neural Network

Objective: To study the superiority and application prospect of generalized regression neural network (GRNN) which is used in forecasting the incidence rate of hypertension. Methods: Use meteorological data, including average temperature, average pressure and relative humidity, and hypertension incidence rate from January 2003 to December 2008 as the input of neural network. Use the incidence rate of hypertension from February 2003 to December 2009 as the output of neural network. Construct the GRNN forecasting model and BP neural network forecasting model respectively with the neural network toolbox of matlab7.0. Fit and forecast the sample and compare the performance between the two different neural networks. Results: The optimize smooth factor of GRNN is 0.24; The hidden layer of BP neural network is 3. From the forecasting effect, the MER between the two neural networks are 0.10% and 0.27% respectively. The MER of GRNN is less than the MER of BP neural network; their R² are 0.956 and 0.867. Conclusion: GRNN is more superior in small sample forecasting than BP neural network, and the forecasting effect is better. GRNN has practical value in solving epidemic problem which has complicate influencing factor such as hypertension.