Synchronous post-exercise electrocardiogram, phonocardiogram, photoplethysmograms and seismocardiogram

Abstract

Noninvasive electromechanical assessment of cardiovascular function is emerging as a cost-effective method for diagnosis of heart failure and arterial diseases, and for telemedical monitoring of blood pressure and neural disorders. It encompasses simultaneous acquisition of electrocardiographic, phonocardiographic, arterial-pulse, chest-vibration, bioimpedance and other waveforms. The phases and amplitudes of these waveforms are used for construction of disease biomarkers. The procedure includes corrections of biomarker values to daily variation and excursions of heart rate. However, datasets that enable a systematic study of the effects of heart rate on mechanical waveforms are currently not available. Here, we describe SensSmartTech - the first dataset of multiparametric cardiovascular signals systematically measured in a large span of heart rates from 52 to 182 beats per minute, achieved by running on a treadmill. Besides providing the data for biomarker correction, the dataset enables new insights into the cardio-respiratory and electro-mechanical couplings in the cardiovascular system.

Fig. 1

(a) Polycardigraph with three modules for synchronized acquisition of ECG, PPG, PCG and ACC signals. The abbreviations used are: LA - left arm, RA - right arm, LL - left leg, RL - right leg and V1-V6 precordial ECG leads, WCT - Wilson central terminal, MCU- microcontroller unit, PWR - power regulators, RLD - right leg drive, LPF - low-pass filter, GPIO - general purpose input/output, I2C - inter-integrated circuit, I2S - inter-integrated circuit sound, SPI - serial peripheral interface, OPA - operational amplifier. (b) Schematics of the sensor synchronization and sampling at the following rates: PCG at 1 kHz, ECG and ACC at 500 Hz and PPG at 100 Hz.

Fig. 2

SensSmartTech sensor positions. (a) An example measurement on a treadmill. (b) Polycardiograph device with ECG, PPG, PCG and ACC sensors and laptop. (c) Schematics of the sensor positions on the body.

Fig. 3

(a) Distribution of median HRs on the whole set. (b) A depiction of the HR relaxation during a typical measurement sequence on one subject. Shown is the instantaneous HR obtained as the inverse of RR interval.

Fig. 4

The study flow showing the number of recruited and rejected subjects, measurement protocol and the number of signals rejected based on different criteria.

Fig. 5

Example polycardiographic recordings. (a), (b) 10-sec snapshots of the polycardiographic recordings taken before running at the median HR of 92 bpm and after running at the median HR of 141 bpm, respectively. (c) Median heartbeats obtained from the whole 30-sec recordings corresponding to (a). Characteristic signal features are shown by black dots. In particular, ECG features - Q, R and S waves of the QRS complex, ${\mathrm{T}}_{\max }$ and T_end points of the T-wave. PPG features: pulse onset (ON), systolic peak (SP), dicrotic notch (DN) and diastolic peak (DP). PCG features: first (S1) and second (S2) heart sounds. ACC features: mitral valve closure (MVC), aortic valve opening (AVO), aortic valve closure (AVC). (d) Median heartbeats obtained from the whole 30-sec recordings corresponding to (a) and (b). Signal amplitudes are given in arbitrary units (au).