A novel algorithm for HRV estimation from short-term acoustic recordings at neck

Abstract

Heart rate variability (HRV) is an important non-invasive parameter to monitor the activity of the autonomicnervous system. This paper proposes an algorithm to analyzeHRV by processing the acoustic data, recorded by placing asmall, wearable sensor on the suprasternal notch (at neck) ofan adult subject, primarily intended to record breathing sounds.The method used an empirical data analysis approach of theHilbert-Huang transform (HHT) to construct an instantaneousenergy envelope and segment the cardiac cycle by detectingS1 and S2 sounds using the K-means algorithm. The time-domain HRV analysis for the short-term recordings of 10subjects demonstrated a close agreement with the referenceECG signal. The instantaneous heart rate (IHR) comparisonsyielded an accuracy of 95.78% and 92.35% for S1 and S2sounds respectively. The experimental results showed that theproposed algorithm can provide an accurate HRV analysis forthe cardiac signals recorded at the neck.