Amplitude and Frequency Based Evaluations for Algorithm Development of Premature Ventricular Contraction Detection System

Heart related diseases are one of the most frequent causes for global death. This reason makes the mitigation and prevention crucial. The system for making the early warning system is developed in this study. The algorithm was written in Python programming language. The inner working for the detection system was based on SciPy module. Premature ventricular complex (PVC) is one of the arrhythmias that must be rapidly detected. Monitoring PVC was important in patients with cardiovascular records as frequent PVC was something need to be aware of in cardiovascular patients. The PVC recognition is developed to detect the abnormality of ECG morphology. The MIT-BIH Arrhythmia Database is used. This dataset supplies ECG data based on the beat and rhythm. The algorithm was able to detect PVC occurrence with average accuracies of 89.31% with standard deviation of 11.26%. However, the algorithm still in its preliminary studies and still contain some limitations in differentiating the PVC and background noise.