Abstract:Computer plays animportant role in every aspect of life. Computers allows huge amount ofstorage, does processing of information at very high speed. Many computersoftware are used to diagnose diseases. Networking makes faster communicationbetween doctor and patient. Digitalstorage of data result in massive amount of data of Patient. It is impossiblefor human to analyze the huge amount of data. Machine learning provides a wayto find out various patterns and reason about the data. Machine learning hasthree models classification, clustering and regression. Various classificationalgorithms are available to classify the data and predict result like Decision tressC4.5, Neural Network MLP, KNN etc. According to the WHO (World HealthOrganization), chronic diseases such as cancer, coronary heart disease,diabetes mellitus type 2, and chronic obstructive pulmonary diseases are amongthe world's most common diseases constitute. Because of this, about 60% of alldeaths occur worldwide. Here, present new health monitoring techniques to theprediction of heart failures. In this system, develop edge-computing basedComplex Event Processing (CEP) techniques with the Remote Patient Monitoring(RPM) for the remote healthcare applications. This approach is based on theCEP, combined with the statistical approach. For the prediction of heartfailure multilayer perceptron (MLP) model is used. The system firstly, collectshealth parameters after that process data using analysis rules. This proposedsystem continuously monitors heart failures patients and after that, itpredicts heart failures strokes based on the related symptoms. When a criticalcondition occurs then it alerts patients. An experimental result shows that theMLP is more accurate than C4.5.
Keywords:-Heart Failures Prediction, C4.5,WHO, Remote Patient Monitoring, and Multilayer Perceptron.