Wavelet-Based Feature Extraction for the Analysis of EEG Signals Associated with Imagined Fists and Feet Movements

Abstract

Electroencephalography (EEG) signals were analyzed in many research applications as a channel of communication between humans and computers. EEG signals associated with imagined fists and feet movements were filtered and processed using wavelet transform analysis for feature extraction. The proposed work used Neural Networks (NNs) as a classifier that enables the classification of imagined movements into either fists or feet. Wavelet families such as Daubechies, Symlets, and Coiflets wavelets were used to analyze the extracted events and then different feature extraction measures were calculated for three detail levels of the wavelet coefficients. Intensive NN training and testing experiments were carried out and different network configurations were compared. The optimum classification performance of 89.11% was achieved with a NN classifier of 20 hidden layers while using the Mean Absolute Value (MAV) of the Coiflets wavelet coefficients as inputs to NN. The proposed system showed a good performance that enables controlling computer applications via imagined fists and feet movements.

Motor Imagery EEG Signal Processing and Classification using Machine Learning Approach

“…Features may be computed for any and corresponding to important MI-based EEG frequency bands. We used coif1 wavelet with level-1 decomposition, as it gives best result among other wavelets [33]. Here, corresponds to beta-band (13-30 Hz) and corresponds to alpha-band (7-13 Hz).…”

EEG Mouse:A Machine Learning-Based Brain Computer Interface

“…A large selection of DWT mother wavelets, such as the Daubechies, Symmlet, and Coif let, is available to be used in our work [22]. But the Coif let(Coif) family of wavelet functions provided the best classification performance in our previous work [11]. So, we decided to calculate the Coif lets wavelets Coif1-Coif5 in this work.…”

“…So, the wavelet transformation of each record at four levels results in four details: cD 1 (40-80Hz), cD 2 (20-40Hz), cD 3 (10-20Hz), and cD 4 (5-10Hz), and a single approximationcA 4 (0-5Hz). As explained in [11], the details cD 2 , cD 3 and cD 4 provided proper representation for the activities of interest. So, we decided to extract the vectors of features from these details only.…”

“…Many amplitude estimators for neurological activities were defined mathematically in [34] and some of them were selected based on our previous results obtained in [11].…”

“…In our previous research [11][12][13] we proposed many systems that could efficiently discriminate between executed (or imagined) left and right fist (or feet) movements. In this work, we integrated these systems into one hybrid application that is based on the imagined fists and feet movements.…”