Efficient Premature Ventricular Contraction Detection Based on Network Dynamics Features

Automatic detection of premature ventricular contractions (PVCs) is essential for early identification of cardiovascular abnormalities and reduction of clinical workload. As the most prevalent arrhythmia, PVCs can cause cardiac failure or sudden death. The difficulty resides in extracting features that effectively reflect the electrocardiogram (ECG) signals. Transition networks (TN), which represent the transition relationships between various phases of a time series, are advantageous for capturing temporal dynamics. Therefore, in order to recognize PVCs, each heartbeat was firstly split into serval segments; then their statistical properties were calculated for the sequence construction; finally, network topology related features were extracted from TN constructed by these sequences of statistical properties, and input into decision trees-based Gentleboost for PVC recognition. The algorithm was trained on MIT-BIH arrhythmia database (MIT-BIH-AR), and tested on St. Petersburg Institute of Cardiological Technics 12-lead arrhythmia database (INCART), wearable ECG database (WECG), and noise stress test database by four evaluation metrics: sensitivity, positive predictivity, F1-score (F1) and area under the curve (AUC). The proposed algorithm achieved an average F1 of 0.9784 and AUC of 0.9975 on MIT-BIH-AR, and proved good generalization ability on INCART and WECG with F1=0.9633 and 0.9467, AUC=0.9887 and 0.9755, respectively. The algorithm also exhibited robustness and noise immunity as evidenced by tests on sensitivity of R-wave peak offset and noise, and real-world daily life conditions. Overall, the proposed PVC detection algorithm based on TN theory offered high classification accuracy, strong robustness, and good generalization ability, with great potential for wearable mobile applications.