Improvement in Neural Respiratory Drive Estimation From Diaphragm Electromyographic Signals Using Fixed Sample Entropy

Improvement in Neural Respiratory Drive Estimation From Diaphragm Electromyographic Signals Using Fixed Sample Entropy 170 177 Journal of Biomedical and Health Informatics (JBHI)

Improvement in Neural Respiratory Drive Estimation From Diaphragm Electromyographic Signals Using Fixed Sample Entropy

Diaphragm electromyography is a valuable technique for the recording of electrical activity of the diaphragm. The analysis of diaphragm electromyographic (EMGdi) signal amplitude is an alternative approach for the indirect quantification of neural respiratory drive (NRD), which reflects the load on the respiratory muscles. The EMGdi signal is, however, corrupted by electrocardiographic (ECG) activity, and this presence of cardiac activity can make the EMGdi interpretation more difficult by medical staff. Traditionally, the EMGdi amplitude has been estimated using the average rectified value (ARV) and the root mean square (RMS). In this work, surface EMGdi signals were analyzed using the fixed sample entropy (fSampEn) algorithm, and compared to traditional ARV and RMS methods. The fSampEn is calculated using a tolerance value fixed and independent of the standard deviation of the analysis window. Thus, this method quantifies the amplitude of the complex components of stochastic signals (such as EMGdi), and being less affected by changes in amplitude due to less complex components (such as ECG). The proposed method was tested in synthetic and recorded EMGdi signals. In different levels of NRD, fSampEn was less sensitive to the effect of cardiac activity on EMGdi signals than ARV and RMS amplitude parameters. Our findings suggest that the proposed method may improve the evaluation of NRD in patients with compromised respiratory function.