Annotation of seismocardiogram using gyroscopic recordings

Chenxi Yang, Sunli Tang, Negar Tavassolian

2016 · 14 citations · 12 references

Concepts

Abstract

This paper introduces a novel setup and algorithm for the automatic annotation of seismocardiographic (SCG) recordings from a MEMS accelerometer. The setup utilizes gyroscopic recordings as a reference for the detection of isovolumic moment (IM) and aortic valve closure (AC) peaks. A method for deriving the rotational kinetic energy waveform is proposed and the coefficients are generated using singular vector decomposition (SVD). Experimental results on 5 subjects at rest indicate an IM detection rate of 96.9% and AC detection rate of 95.6% without envelope filtering. It is suggested that this algorithm is feasible as an ECG-free automatic peak annotation method of SCG recordings from subjects at rest.

References

12