Neural and Mathematical Predicting Models For Particulate Matter Impact on Human Health in Oman

The recorded reports of the World Health Organization (WHO) show that a total of 4.2 million death cases is due to exposure to PM2.5 particulate matter. This paper aims to analyze and examine the impact of particulates (PM 2.5 and PM10) on human health in Oman. Also, it proposed neural and mathematical prediction models for predicting predict the future levels of particulate matter (PM2.5 and PM10) and its influence on human health. The paper performs a critical comparative study of proposed models, which is evident that the proposed models were fast, cheap, and accurate. The first model is based on Linear regression that obtained results of the coefficient of determination R² =0.7604, mean square error (MSE=0.0673), and root mean square error (RMSE=0.2595). The second model is based on non-linear regression polynomial that achieved excellent results of (R²) value of 0.9394 and (MSE) value of 0.0209 and (RMSE) value of 0.1447. The Neural model is more accurate in predicting the experimental results, which is obtained the highest achievements of MSE value =0.0064, correlation rate (R) =0.994, and NMSE =0.01392. The work confirmed that the Arab countries and Oman in a good and moderate situation based AQI indicator and did not reach the degree of danger of pollutants