ECG classification through complexity measures using random forests

Aging causes changes in the cardiovascular structure, increasing the risk of diseases, and making people more dependent and vulnerable. According to the National Institute of Social Security, there are approximately 250,000 new cases of stroke in Brazil every year, and around 40% of retirement requests are due to heart attacks or strokes. Heart diseases are one of the most prominent causes of death in the world. Most sudden death cases occur without previous symptoms, whereas some non-lethal types of arrhythmias such as ventricular extrasystoles precede other directly related cases. In this context, it is advisable to monitor high-risk individuals, who are not hospitalized daily. Considering the aging population and the increasing number of people living alone, it is important to monitor for various types of biomedical signals remotely. This study aims to evaluate and compare three approaches to the QRS classification using the MIT-BIH Arrhythmia Database.