Arrhythmia is one of the major cause of deaths across the globe. Almost 17.9 million deaths are caused due to cardiovascular diseases. In order to reduce this much mortality rate, the cardiovascular disease should be properly identified and the proper treatment for the same should be immediately provided to the patients. In this study, a new ensemble based support vector machine (SVM) classifier was proposed to classify heartbeat into four classes from MIT-BIH arrhythmia database. The results were compared with other classifiers that are SVM, Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short Term Memory network. The four features were extracted from the ECG signals that were used by the classifiers are Wavelets, high order statistics, R-R intervals and morphological features. An ensemble of SVMs obtained the best result with an overall accuracy of 94.4%.
The prime cause of deaths across the world according to the report of WHO are cardiovascular diseases (CVDs): Annually, more people die from cardiac diseases than from any other disease. In 2016, approximately 18 million people died because of cardiac-related diseases, which reflects 31% of all the deaths globally. 85% of the deaths among this 31% are due to heart attack and stroke. An approximate of three-fourths of cardiac deaths take place in low-income and middle-income countries . In 2015, 82% of the 17 million premature deaths due to non- communicable diseases are in low and middle-income countries, and the rest are caused by CVDs. The leading cause of CVDs is a long-term effect of cardiac arrhythmias. When the electrical signal, to the heart that coordinate heartbeats don’t work properly, Arrhythmias occur .
Arrhythmia is a condition in which heartbeat is either irregular, too fast or too slow. Most of the heart arrhythmias are generally not harmful; In case if they seem to be exceptionally abnormal, or result from a weak or damaged heart, arrhythmias can even cause serious and potentially fatal symptoms. The electrocardiogram (short termed as ECG) is a significant diagnostic tool that is used to assess and monitor the electrical activities and muscular functions of the human heart . These heart activities result in creating some waves called P-QRS-T waves (see Fig. 1). Even though, it is a really simple test to carry on, the examination and determination of the ECG tracing require huge amounts of training. It’s not mandatory to have symptoms for all the arrhythmias. Some arrhythmias do exist with no sort of symptoms. For those arrhythmias that exhibit symptoms, some of the following may be symptoms: dizziness, breathlessness and noticeably rapid, strong, or irregular heartbeat due to agitation.
This paper used MIT-BIH arrhythmia patient-specific dataset to classify the heartbeats into 4 classes which include one normal beat and 3 abnormal beats i.e. SVEB, VEB and F. Data pre-processing tasks were performed which includes baseline noise removal, heartbeat segmentation which segments the heartbeats within window size of 180 centered at R-peak and normalization using Z-score. Then important features were extracted namely R-R intervals, HOS (high order statistics), Wavelets and morphological features.
Further, classification was done using five techniques which are Support vector machines (SVM), random forest (RF), long short term memory (LSTM), K-nearest neighbours (KNN) and ensemble of SVMs. The best result was obtained using ensemble of SVMs with overall accuracy of 94.4%, average accuracy of 97.2%, sensitivity of 65.26%, specificity of 93.35%, precision of 69.11% and F-score of 66.24%. Among the single classifiers, Random forest performed good with overall accuracy of 93.25% and mean accuracy of 96.73%. In this paper the potential of ensembling has been unwind and has been proven best among all the single classifiers.