Automatic Classification of Heartbeats Using ECG Morphology and Heartbeat Interval Features

Abstract

Abstract-A method for the automatic processing of the electrocardiogram (ECG) for the classification of heartbeats is presented. The method allocates manually detected heartbeats to one of the five beat classes recommended by ANSI/AAMI EC57:1998 standard, i.e., normal beat, ventricular ectopic beat (VEB), supraventricular ectopic beat (SVEB), fusion of a normal and a VEB, or unknown beat type. Data was obtained from the 44 nonpacemaker recordings of the MIT-BIH arrhythmia database. The data was split into two datasets with each dataset containing approximately 50 000 beats from 22 recordings. The first dataset was used to select a classifier configuration from candidate configurations. Twelve configurations processing feature sets derived from two ECG leads were compared. Feature sets were based on ECG morphology, heartbeat intervals, and RR-intervals. All configurations adopted a statistical classifier model utilizing supervised learning. The second dataset was used to provide an independent performance assessment of the selected configuration. This assessment resulted in a sensitivity of 75.9%, a positive predictivity of 38.5%, and a false positive rate of 4.7% for the SVEB class. For the VEB class, the sensitivity was 77.7%, the positive predictivity was 81.9%, and the false positive rate was 1.2%. These results are an improvement on previously reported results for automated heartbeat classification systems.Index Terms-Electrocardiogram (ECG), heartbeat classifier, linear discriminant analysis, statistical classifier model.

A novel method of diagnosing premature ventricular contraction based on sparse auto-encoder and softmax regression

“…In recent years, many studies concerning the classification of arrhythmias have been conducted [4][5][6][7][8][9][10][11][12]. In one study, J. Wang developed a novel ECG arrhythmia classification method based on feature reduction by combing a principal component analysis (PCA) with a linear discriminant analysis (LDA).…”

“…Peng Li developed a low complexity data-adaptive PVC recognition approach that exhibited good robustness against noise, generalization capabilities, and a PVC recognition accuracy of 98.2%, indicating that it could be effectively used for real-time applications [16]. Using these algorithms, the features of ECGs were manually extracted based on time domain information, such as ECG morphology [6,7,11,12], and transform domain information [4,5,9,[12][13][14], such as the wavelet transforms or statistical parameters [10,15,16]. These processes require artificial experience or specialized knowledge and increase computational complexity.…”

Untitled

“…Perbandingan beberapa metode ekstraksi fitur seperti PCA, transformasi wavelet, algoritma morfologi yang diimplementasikan dalam algoritma klasifikasi berbasis JST [15] dan SVM [16]. Ekstraksi fitur mengunakan algoritma morfologi juga dilakukan oleh Philips dan kawankawan [17]. Namun dari sekian banyak penelitian yang telah dilakukan oleh para peneliti dalam klasifikasi beat, masih jarang dijumpai adanya topik yang mengembangkan suatu algoritma yang mampu untuk mendeteksi adanya outlier klasifikasi.…”

Feature Selection for Inter-Patient Supervised Heart Beat Classification

“…In the case of the LDA classifier, the following decomposition and weighting is applied to the sum over observations in (1) [4]:…”