Cough detection through mechanomyographic signal in synchronized respiratory electrical stimulation systems.

Abstract

Synchronization of transcutaneous functional electrical stimulation (TFES) with the spontaneous inspiration and expiration phases is a new approach for respiratory rehabilitation. One of the requirements for proper operation is the identification of cough events to automatically change the stimulation parameters in order to increase muscle strength during the cough. The aim of this work is to assess the viability in detection of cough events with a mechanomyographic (MMG) sensor on the abdominal region, and to evaluate if it can be used simultaneously with the synchronized TFES system. An MMG sensor was placed in contact with skin lined with the last ribs, above the rectus abdominis muscle and the linea alba. Two tests were accomplished which included quiet breathing, speaking and coughing episodes. The developed system efficiently distinguishes quiet breathing and coughing signals, but speaking is still confused with coughing episodes. The MMG sensor suffered detectable amplitude changes mainly during the forced expiration phase of the cough, but it could also detect the compression phase at lower amplitude. Therefore, the MMG system can be used for cough detection in this application.