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 quantification of the neural respiratory drive (NRD). The EMGdi signal is, however, corrupted by electrocardiographic (ECG) activity, and this presence of cardiac activity can make the EMGdi interpretation more difficult. Traditionally, the EMGdi amplitude has been estimated using the average rectified value (ARV) and the root mean square (RMS). In this study, surface EMGdi signals were analyzed using the fixed sample entropy (fSampEn) algorithm, and compared to the 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. fSampEn was less sensitive to the effect of cardiac activity on EMGdi signals with different levels of NRD than ARV and RMS amplitude parameters. The mean and standard deviation of the Pearson's correlation values between inspiratory mouth pressure (an indirect measure of the respiratory muscle activity) and fSampEn, ARV, and RMS parameters, estimated in the recorded EMGdi signal at tidal volume (without inspiratory load), were 0.38 ± 0.12, 0.27 ± 0.11, and 0.11 ± 0.13, respectively. Whereas at 33 cmH ₂ O (maximum inspiratory load) were 0.83 ± 0.02, 0.76 ± 0.07, and 0.61 ± 0.19, respectively. Our findings suggest that the proposed method may improve the evaluation of NRD.

The analysis of amplitude parameters of the diaphragm mechanomyographic (MMGdi) signal is a non-invasive technique to assess respiratory muscle effort and to detect and quantify the severity of respiratory muscle weakness. The amplitude of the MMGdi signal is usually evaluated using the average rectified value or the root mean square of the signal. However, these estimations are greatly affected by the presence of cardiac vibration or mechanocardiographic (MCG) noise. In this study, we present a method for improving the estimation of the respiratory muscle effort from MMGdi signals that is robust to the presence of MCG. This method is based on the calculation of the sample entropy using fixed tolerance values (fSampEn), that is, with tolerance values that are not normalized by the local standard deviation of the window analyzed. The behavior of the fSampEn parameter was tested in synthesized mechanomyographic signals, with different ratios between the amplitude of the MCG and clean mechanomyographic components. As an example of application of this technique, the use of fSampEn was explored also in recorded MMGdi signals, with different inspiratory loads. The results with both synthetic and recorded signals indicate that the entropy parameter is less affected by the MCG noise, especially at low signal-to-noise ratios. Therefore, we believe that the proposed fSampEn parameter could improve estimates of respiratory muscle effort from MMGdi signals with the presence of MCG interference.

In this work, it is proposed a technique for the feature extraction of electroencephalographic (EEG) signals for classification of mental tasks which is an important part in the development of Brain Computer Interfaces (BCI). The Empirical Mode Decomposition (EMD) is a method capable to process nonstationary and nonlinear signals as the EEG. This technique was applied in EEG signals of 7 subjects performing 5 mental tasks. For each mode obtained from the EMD and each EEG channel were computed six features: Root Mean Square (RMS), Variance, Shannon Entropy, Lempel-Ziv Complexity Value, and Central and Maximum Frequencies, obtaining a feature vector of 180 components. The Wilks' lambda parameter was applied for the selection of the most important variables reducing the dimensionality of the feature vector. The classification of mental tasks was performed using Linear Discriminate Analysis (LD) and Neural Networks (NN). With this method, the average classification over all subjects in database was 91plusmn5% and 87plusmn5% using LD and NN, respectively. It was concluded that the EMD allows getting better performances in the classification of mental tasks than the obtained with other traditional methods, like spectral analysis.

Fixed sample entropy (fSampEn) is a robust technique that allows the evaluation of inspiratory effort in diaphragm electromyography (EMGdi) signals, and has potential utility in sleep studies. To appropriately estimate respiratory effort, fSampEn requires the adjustment of several parameters. The aims of the present study were to evaluate the influence of the embedding dimension m, the tolerance value r, the size of the moving window, and the sampling frequency, and to establish recommendations for estimating the respiratory activity when using the fSampEn on surface EMGdi recorded for different inspiratory efforts. Values of m equal to 1 and r ranging from 0.1 to 0.64, and m equal to 2 and r ranging from 0.13 to 0.45, were found to be suitable for evaluating respiratory activity. fSampEn was less affected by window size than classical amplitude parameters. Finally, variations in sampling frequency could influence fSampEn results. In conclusion, the findings suggest the potential utility of fSampEn for estimating muscle respiratory effort in further sleep studies.

Epilepsy is a neurological disorder that affects around 50 million people worldwide. The seizure detection is an important component in the diagnosis of epilepsy. In this study, the Empirical Mode Decomposition (EMD) method was proposed on the development of an automatic epileptic seizure detection algorithm. The algorithm first computes the Intrinsic Mode Functions (IMFs) of EEG records, then calculates the energy of each IMF and performs the detection based on an energy threshold and a minimum duration decision. The algorithm was tested in 9 invasive EEG records provided and validated by the Epilepsy Center of the University Hospital of Freiburg. In 90 segments analyzed (39 with epileptic seizures) the sensitivity and specificity obtained with the method were of 56.41% and 75.86% respectively. It could be concluded that EMD is a promissory method for epileptic seizure detection in EEG records.