Wavelet transform feature extraction from human PPG, ECG, and EEG signal responses to ELF PEMF exposures: A pilot study

Extraction of Various Features of ECG Signal

“…The human heart is located in the chest between the lungs, behind the sternum and above the diaphragm. It weighs between 200 to 425 grams and is a little larger than the size of a fist [2,3,4]. The basis end and the apex end of the heart lie on its main axis which is oriented from the back-topright to the front-bottom-left of the torso.…”

Identification of ERD using Fuzzy Inference Systems for Brain-Computer Interface

“…In [28] a novel method was proposed for the feature extraction of electroencephalogram (EEG) based on wavelet packet decomposition (WPD). In [29] the assessment of wavelet transform (WT) as a feature extraction method was used in representing the electrophysiological signals.…”

Time-Frequency Based Feature Extraction for Non-Stationary Signal Classification

“…-Regional analysis approaches (Bernat et al, 2007;Cvetkovic et al, 2008;Hassanpour et al, 2004;Subasi, 2007;Tzallas et al, 2008) In this approach, the TFR is decomposed in regions of interest and further analysis is carried out with respect to the behavior on each of such regions defined on the t-f plane (Tzallas et al, 2008). For example, a partition of the t-f plane can be regarded, defining a grid depending on some ap r i o r iinformation known about the information distribution over the plane.…”

“…This method is less computationally expensive compared to distance based approaches, but an adequate selection of the regions of interest should be attained in order to achieve accurate results. Other regional decomposition can be accomplished by transform methods as proposed in (Bernat et al, 2007;Hassanpour et al, 2004) or wavelet sub-band decompositions (Cvetkovic et al, 2008;Subasi, 2007).…”

“…In order to make pattern recognition problems solvable, it is necessary to convert t-f patterns into feature vectors, which become condensed representations, ideally containing only relevant information. From the viewpoint of managing large quantities of data, it is even more useful if irrelevant or redundant attributes could be segregated from relevant and important ones (Cvetkovic et al, 2008). A direct approach is to use linear transform methods, such as principal component analysis (PCA) or partial least squares (PLS), to reduce the feature space dimensionality resulting from the t-f plane.…”