Multi-objective optimization for EEG channel selection and accurate intruder detection in an EEG-based subject identification system

Abstract

We present a four-objective optimization method for optimal electroencephalographic (eeG) channel selection to provide access to subjects with permission in a system by detecting intruders and identifying the subject. Each instance was represented by four features computed from two subbands, extracted using empirical mode decomposition (eMD) for each channel, and the feature vectors were used as input for one-class/multi-class support vector machines (SVMs). We tested the method on data from the event-related potentials (ERPs) of 26 subjects and 56 channels. The optimization process was performed by the non-dominated sorting genetic algorithm (NSGA), which found a threechannel combination that achieved an accuracy of 0.83, with both a true acceptance rate (TAR) and a true rejection rate (TRR) of 1.00. In the best case, we obtained an accuracy of up to 0.98 for subject identification with a TAR of 0.95 and a TRR 0.93, all using seven EEG channels found by NSGA-III in a subset of subjects manually created. The findings were also validated using 10 different subdivisions of subjects randomly created, obtaining up to 0.97 ± 0.02 of accuracy, a TAR of 0.81 ± 0.12 and TRR of 0.85 ± 0.10 using eight channels found by NSGA-III. These results support further studies on larger datasets for potential applications of EEG in identification and authentication systems.

Towards a minimal EEG channel array for a biometric system using resting-state and a genetic algorithm for channel selection

“…We previously showed that the TAR and TRR of the models created using OC-SVM can be improved by finding the best nu and gamma parameters 12 . We performed the optimization process defined in the "Methods" section to provide more information about the behavior of OC-SVM models using a bigger dataset, trying to improve the TAR and TRR while reducing the necessary number of EEG channels for subject identification.…”

“…For EEG-based biometric systems, several approaches have been presented using various paradigms to stimulate and record the EEG signals, i.e. imagined speech [1][2][3] , resting-state [4][5][6][7][8][9][10] , and event-related potentials (ERPs) 11,12 .…”

“…In general, there are a number of approaches that use raw data as input for various configurations of neural networks (NN) [13][14][15][16] and several have been proposed to tackle the high dimensionality of the data using methods for feature extraction, such as principal component analysis (PCA) 17 , the discrete wavelet transform (DWT) 3 , or empirical mode decomposition (EMD) 2,3,11,12 .…”

“…Based on the results of our previous studies 2,7,12 , we compared EMD and DWT for decomposing the EEG signals into different frequency bands. When using EMD, we selected the two sub-bands closest to the original EEG signal for each channel, and for DWT we decomposed the EEG signal into four different sub-bands (Three levels of decomposition and the residual), as explained in the "Methods" section.…”

“…We solved the optimization problem using the non-dominated sorting genetic algorithm (NSGA)-III, which uses the Pareto-optimal solutions and a niche method to maintain stable sub-populations of good points 37 . NSGA-III has proven to be a robust method for channel selection, especially when using more than two objectives 12,38 . First, we report the results for a set of experiments using all the available channels to show the TAR and TRR using different parameters.…”