Uma plataforma de monitoramento inteligente de arritmia cardíaca em fluxo de tempo real

Abstract

Arrhythmia affects millions of people. In Europe and North America, since 2014, atrial fibrillation affects about 2% to 3% of the population. One of the ways to detect arrhythmias is by performing an electrocardiographic examination is an ECG machine that records the electrical activity of the heart and displays this data as a trace on a paper, then these data are interpreted by a physician and thus given the diagnosis. Thus, it was thought a step further in assisting the physician with the automatic diagnosis of cardiac arrhythmia. Thus, this study aims to plan and develop a smart platform capable of automatically monitoring and identifying cardiac arrhythmias and notifying health professionals, family members and patients in real-time flow. Therefore, we used base classifiers and classifying committees, such as: AdaBoost, RF J48, kNN, MLP, NB and SVM.