This paper proposes a comprehensive study on machine listening for localisation of snore sound excitation. Here we investigate the effects of varied frame sizes, and overlap of the analysed audio chunk for extracting low-level descriptors. In addition, we explore the performance of each kind of feature when it is fed into varied classifier models, including support vector machines, k-nearest neighbours, linear discriminant analysis, random forests, extreme learning machines, kernel-based extreme learning machines, multilayer perceptrons, and deep neural networks. Experimental results demonstrate that, wavelet packet transform energy can outperform most other features. A deep neural network trained with subband energy ratios reaches the highest performance achieving an unweighted average recall of 72.8% from four types for snoring.
Electroencephalogram (EEG) is one of biomedical signals measured during all-night polysomnography to diagnose sleep disorders, including sleep apnoea. Usually two central EEG channels (C3-A2 and C4- A1) are recorded, but typically only one of them are used. The purpose of this work was to compare discriminative features characterizing normal breathing, as well as obstructive and central sleep apnoeas derived from these central EEG channels. The same methodology of feature extraction and selection was applied separately for the both synchronous signals. The features were extracted by combined discrete wavelet and Hilbert transforms. Afterwards, the statistical indexes were calculated and the features were selected using the analysis of variance and multivariate regression. According to the obtained results, there is a partial difference in information contained in the EEG signals carried by C3-A2 and C4-A1 EEG channels, so data from the both channels should be preferably used together for automatic sleep apnoea detection and differentiation.
EEG signal-based sleep stage classification facilitates an initial diagnosis of sleep disorders. The aim of this study was to compare the efficiency of three methods for feature extraction: power spectral density (PSD), discrete wavelet transform (DWT) and empirical mode decomposition (EMD) in the automatic classification of sleep stages by an artificial neural network (ANN). 13650 30-second EEG epochs from the PhysioNet database, representing five sleep stages (W, N1-N3 and REM), were transformed into feature vectors using the aforementioned methods and principal component analysis (PCA). Three feed-forward ANNs with the same optimal structure (12 input neurons, 23 + 22 neurons in two hidden layers and 5 output neurons) were trained using three sets of features, obtained with one of the compared methods each. Calculating PSD from EEG epochs in frequency sub-bands corresponding to the brain waves (81.1% accuracy for the testing set, comparing with 74.2% for DWT and 57.6% for EMD) appeared to be the most effective feature extraction method in the analysed problem.