A portable multisensor system to assess cardiorespiratory interactions through photoplethysmography

Nowadays, the ever-growing interest to health and quality of life of individuals and the advancements in electronic devices technology are pushing the development of portable and wearable biomedical devices able to pursue a minimally invasive monitoring of physiological parameters in daily-life conditions. Such devices can now carry out a real-time assessment of the subjects' overall health status and possibly even detect ongoing diseases. In this context, we have designed and implemented a multisensor portable system able to perform synchronous real-time acquisitions of electrocardiographic (ECG), photoplethysmographic (PPG) and airflow breathing signals. We investigated cardiorespiratory interactions between heart period and respiratory time series, extracted from combined ECG and breathing signals (considered as the reference), or using the PPG signal only, through Granger Causality measures in time and frequency domain. The aim was to assess to what extent the non-invasive and cost-effective PPG technique can be employed alone to assess cardiorespiratory interactions, thus avoiding the simultaneous acquisitions of ECG or breathing signals with more bulky or uncomfortable devices. The analysis was carried out on 6 healthy young subjects, undergoing a two-phase protocol consisting in spontaneous and controlled breathing phases. Our findings show that linear interactions measures behave similarly if ECG or PPG are used for detecting the heart period and sampling the airflow respiratory signal, while the utilization of a respiratory signal extracted through filtering or as the envelope of the PPG waveform could lead to causality underestimates and must be further investigated.