Brain-Computer Interface for wheelchair control operations: An approach based on Fast Fourier Transform and On-Line Sequential Extreme Learning Machine

Abstract

Objective: The paper aims to control an electric wheel chair with a Brain-Computer Interface (BCI) headset. This wheel chair would be helpful for disabled people who cannot move their hands and legs or basically suffering from cerebromedullospinal disconnection. The main objective is to map different facial expression to the movement of the wheelchair. Methods: The headset used for this purpose comprises of an EEG cap which has a total of 16 electrodes connected out of which 14 electrodes are used for acquiring data and the rest 2 are used as ground and reference. To operate the wheelchair the subject has to place the cap on his/her head and different facial expressions (clench, smile, blink, etc.) are performed. The movement of muscles around the face due to facial expressions can be observed from the EEG signals being recorded. Each expression is linked to movement of the wheelchair. Signal is preprocessed to remove the artifacts and then features are extracted using Fast Fourier Transform. In this paper, for classification we have used state-of-the-art machine learning technique which is a slight variation of Extreme Learning Machine called as Online Sequential Extreme Learning Machine abbreviated as OS-ELM.

Determining threshold value on information gain feature selection to increase speed and prediction accuracy of random forest

“…Several preliminary studies examined feature selection for Random Forest. For instance, Yunming's [28] study proposed a stratified sampling method to select feature subspace for Random Forest with high dimensional data as well as strong and weak informative features.…”

“…The transformed dataset is returned using IFFT. In addition, other studies used the FFT algorithm to perform feature extraction [28,29,37]. For instance, Ansari used Fast Fourier Transform (FFT) to extract features in the EEG dataset [28], with better accuracy than other classifiers.…”

“…Therefore, based on the results of these studies, FFT is applied in this study. Besides being applied to the dataset, FFT is also applied to feature extraction [28,29], and selection [30]. Gowid et al [30] used used this process to develop robust, fast, and automated feature selection algorithms for mechanical systems.…”

Multi-Sessions Outcome for EEG Feature Extraction and Classification Methods in a Motor Imagery Task

“…The most successful feature extraction methods widely used are: common spatial pattern [13], Fourier transform [14], independent component analysis [15], phase synchronization [16], wavelet transform [17] and autoregressive (AR) spectral estimation [18].…”

Determining Threshold Value on Information Gain Feature Selection to Increase Speed and Prediction Accuracy of Random Forest

“…This data can be presented in a continuous wave form, so that it can be presented briefly by using the FFT algorithm [19]. Other researchers also used the FFT algorithm to perform feature extraction [20], [21]. Pratama, in his research stated that real and imaginary numbers resulted from the FFT calculation can be used for the feature selection algorithm [22].…”