Influence of torso movements on a multi-sensor garment for respiratory monitoring during walking and running activities

Unobtrusive and wearable technologies are gaining broad acceptance in the continuous monitoring of physiological parameters. Among others, the respiratory frequency is increasingly being considered as it allows to detect physiological abnormalities and health status changes, both in clinical and occupational scenarios and in sports science.The respiratory monitoring during physical activity is still challenging because of the artifacts caused by body movements. These breathing-unrelated movements may negatively affect the signal used for respiratory monitoring.This study aimed to investigate the influence of the torso movements that occur during walking and running activities on the signals recorded by a multi-sensor garment. The garment consists of three bands positioned at the level of upper and lower thorax and abdomen. Each band embeds two conductive sensors. The torso movements were recorded by a magneto-inertial measurement unit embedded into the garment.Experimental trials were carried out on four male volunteers during walking and running activities at selected speeds controlled by a treadmill. A flowmeter was used to retrieve reference respiratory frequency values. All the signals were analyzed in the frequency domain to investigate the frequency content.Results show that movements related to the torso rotation (arms and shoulders swing) cause motion artifacts on the garment sensors’ signals. Sensors positioned on the upper thorax and abdomen were found to be much more influenced by breathing-unrelated movements.Despite the influence of the torso movements on the conductive sensors, the garment can provide robust information about the average respiratory rate even during physical activities.