Development of a software tool for automated surface EMG analysis of respiratory muscles during mechanical ventilation

Photo credit – Shutterstock

The diaphragm is the most important muscle of the respiratory pump. The condition of the human diaphragm can be analyzed via a so-called electromyogram (EMG). This EMG can be recorded with dedicated electrodes and represents its electrical muscular activity and thereby allowing for the monitoring of the patient’s diaphragmatic condition. Two clinical domains are identified where diaphragmatic monitoring is of crucial importance: Intensive Care and home mechanical ventilation. The current gold standard to measure the electrical activity of the diaphragm is the insertion of a nasogastric tube mounted with electrodes. However, this is invasive, data acquisition is cumbersome and analysis remains time-consuming and complex. For both clinical domains, this is far from ideal, as patients are vulnerable. Alternatively, diaphragmatic EMGs recorded via superficial skin electrodes is a novel, promising approach, which is easy to use and noninvasive. These surface electrodes measure, however, not only diaphragm electrical activity, but also noise from other skeletal muscles and the heart. This outlines the clinical challenges in the analyses of diaphragm activity, which can significantly be improved based on current insights and further expanded to provide clinically very useful additional information readily contained in the acquired signals. We, therefore, propose to develop a software tool for (online) automated noise and artifact reduction and analysis of existing and also new parameters derived from surface EMG data. Improved and more extensive analysis of diaphragmatic EMG will result in better research and patient care for individuals requiring mechanical ventilation in both the Intensive Care or home mechanical ventilation domain.

Participating organisations

University of Twente
Life Sciences
Life Sciences
Netherlands eScience Center

Impact

Output

Team

Candace Makeda  Moore
Candace Makeda Moore
Research Software Engineer
Netherlands eScience Center
EO
Eline Oppersma
Principal investigator
University of Twente

Related projects

Diagnosis of active epilepsy in resource-poor setting

Prediction models based on EEG characteristics

Updated 18 months ago
Finished

Related software

ReSurfEMG

RE

ReSurfEMG is an open source collaborative python library for analysis of respiratory electromyography (EMG). The library includes functions for pre-processing, analysis and machine learning.

Updated 21 months ago

ReSurfEMG-dashboard

R

A dashboard for visualizing and analyzing respiratory EMG signal, making use of the ReSurfEMG library

Updated 17 months ago
2