Elman Neural Network Mortality Predictor for Prediction of Mortality Due to Pollution

Air pollution is a significant risk factor for a number of health conditions including respiratory infections, heart disease like stroke and lung cancer. Further, air pollution exposure is a risk factor correlating with increased total mortality from cardiovascular events and Lung disease such as chronic bronchitis and emphysema. A severe health issue is on the other hand constituted by high levels of pollutants associated in the epidemiological literature with an increase in the mortality and cardio respiratory hospitalizations. In this research work, an Elman network is developed to investigate the association between air pollution and the mortality and to predict mortality due to pollution. The Elman Neural Network Mortality Predictor (ENNMP) predicts the respiratory mortality, cardiovascular mortality and the total mortality from the input data set containing the key variables of Temperature, Relative humidity, Carbon monoxide, Sulphur dioxide, Nitrogen dioxide, Hydrocarbons, Ozone and Particulates (PM). The advantage of the usage of neural network for prediction is that they are able to learn from examples only and that after their learning is finished, they are able to catch hidden and strongly non-linear dependencies, even when there is a significant noise in the training set. Dynamic neural networks are good at time series prediction. Elman neural model being recurrent neural network, architecture explores the prediction process with its faster convergence. The simulation results are presented and analysed. The results confirm that the utility of Elman neural network for the prediction of mortality with good accuracy.