EEG signal improvement with cascaded filter based on OWA operator

Abstract

Noise reduction methods have a great impact on the performance of all EEG signal processing systems. The work involves the reduction in impulsive interferences. It is impossible to suppress effectively such a noise using linear filtering approach. The work presents the properties of a cascaded filter based on the ordered weighted aggregation (OWA) operator and its application to improve the electroencephalogram (EEG) signal. The OWA operator is a class of mean-like aggregation operators. By introducing a nonlinear sorting process and by assigning appropriate values of weights to the sorted samples, the improvement in filtering in the impulsive environment is achieved. The structure of the proposed cascade consists of two layers. The first layer contains two OWA filters that share a certain number of signal samples. The second layer averages the outputs of the first layer. The algorithm for the rise of filtering efficiency has been introduced. The performance of the new method has been experimentally compared with the traditional methods of impulsive noise reduction, using synthetic as well as real signals from the CHB-MIT database. The obtained results demonstrate that in the field of impulsive noise suppression and preprocessing, the proposed cascaded OWA filter brings about a significant improvement in the signal-to-noise ratio.

Comparison of Smoothing Filters in Analysis of EEG Data for the Medical Diagnostics Purposes

“…As mentioned above, Savitzky-Golay is a digital polynomial filter (or a least smoothing filter) [25]. Both filters are smoothing filters [26][27][28]. The classic, 'basic' smoothing filter smooths the data in the column vector using a moving average filter, which works in the way that it replaces each data point with the average of the neighbor data points (defined within its span).…”

“…This is because smoothing of bio-medical signals require additional attention as the data, in particular, EEG signals, are very sensitive and prone to various artifacts. Some frequency ranges may also contain crucial information, potentially important for diagnostic purposes, and an incorrect choice of processing or filtering methods may affect these [27].…”

“…This work discusses the advantages and disadvantages of classic smooth filters such as medfilter or Savitzky-Golay filter, which were chosen because they differ from other filters as they reduce the risk of data cutouts, and the smoothed values can be written as a linear transformation of the values. The performed smoothing operation is known as a linear smoother [25,27].…”

“…In practice, this means that they retain the so-called edges, which are frequently present in the EEG signals [36][37][38]40,41]. Smoothing filters' (Savitzky-Golay and Median filters) application enables to correct inter alia spikes present in the data [25,27].…”

“…The more neighbors are applied, the smoother will be the final signal [25]. It smooths the fluctuations and increases the signal-noise ratio (SNR) without significant distortion of the analyzed data [27,42].…”