Wearable seismocardiography: towards a beat-by-beat assessment of cardiac mechanics in ambulant subjects.

Seismocardiogram (SCG) is the measure of the micro-vibrations produced by the heart contraction and blood ejection into the vascular tree. Over time, a large body of evidence has been collected on the ability of SCG to reflect cardiac mechanical events such as opening and closure of mitral and aortic valves, atrial filling and point of maximal aortic blood ejection. We recently developed a smart garment, named MagIC-SCG, that allows the monitoring of SCG, electrocardiogram (ECG) and respiration out of the laboratory setting in ambulant subjects. The present pilot study illustrates the results of two different experiments performed to obtain a first evaluation on whether a dynamical assessment of indexes of cardiac mechanics can be obtained from SCG recordings obtained by MagIC-SCG. In the first experiment, we evaluated the consistency of the estimates of two indexes of cardiac contractility, the pre-ejection period, PEP, and the left ventricular ejection time, LVET. This was done in the lab, by reproducing an experimental protocol well known in literature, so that our measures derived from SCG could have been compared with PEP and LVET reference values obtained by traditional techniques. Six healthy subjects worn MagIC-SCG while assuming two different postures (supine and standing); PEP was estimated as the time interval between the Q wave in ECG and the SCG wave corresponding to the opening of aortic valve; LVET was the time interval between the SCG waves corresponding to the opening and closure of the aortic valve. The shift from supine to standing posture produced a significant increase in PEP and PEP/LVET ratio, a reduction in LVET and a concomitant rise in the LF/HF ratio in the RR interval (RRI) power spectrum. These results are in line with data available in literature thus providing a first support to the validity of our estimates. In the second experiment, we evaluated in one subject the feasibility of the beat-by-beat assessment of LVET during spontaneous behavior. The subject was continuously monitored by the smart garment from 8 am to 8 pm during a workday. From the whole recording, three data segments were selected: while the subject was traveling to work (M1), during work in the office (O) and while traveling back home (M2). LVET was estimated on a beat-by-beat basis from SCG and the RRI influence was removed by regression analysis. The LVET series displayed marked beat-by-beat fluctuations at the respiratory frequency. The amplitude of these fluctuations changed in the three periods and was lower when the LF/HF RRI power ratio was higher, at O, thus suggesting a possible influence of the autonomic nervous system on LVET short-term variability. To the best of our knowledge this case report provides for the first time a representation of the beat-by-beat dynamics of a systolic time interval during daily activity. The statistical characterization of these findings remains to be explored on a larger population.