Standalone Heartbeat Extraction in SCG Signal Using Variational Mode Decomposition

Abstract

In this paper, a variational mode decomposition (VMD) based heartbeat extraction framework is proposed for seismocardiogram (SCG) signal. A reference cardiac signal such as ECG is not needed in our proposed method. The proposed method consists of four major steps: signal decomposition using VMD algorithm, heart rate (HR) envelope construction, low pass filtering of constructed envelope, and annotation of smoothed envelope. The method annotates the HR envelope in terms of characteristic points such as PZCI, NZCI, PI, and TI. Each of the characteristic points can be used for SCG cycle extraction. The proposed method is tested and validated with CEBS database available at the Physionet archieve. Based on the experimental results, it is observed that the proposed method with peak instances (PI) achieves consistent results with good accuracy among all. The qualitative analysis of performance results shows good performance of the proposed method for healthy subjects.