Complex event processing of health data in real time predicting heart failure risk

Abstract

In this article, we develop a scalable system that can perform heart failure prediction techniques based on complex event processing (CEP). The emergence of different health conditions can be seen as complex events and therefore this concept can be easily extended to other uses. The system uses MLP (Multilayer Perceptron) for the prediction of heart failure. First, perform preprocessing and after that collect the health parameter. The system monitors the patients of heart failure and predicts heart attacks. When critical conditions are occurs the system warns the patients. Experimental results show that MLP is more accurate than C 4.5, based on Precision-Recall and F1.