Visual Object Recognition From Single-Trial EEG Signals Using Machine Learning Wrapper Techniques

Automated Classification of Cognitive Visual Objects Using Multivariate Swarm Sparse Decomposition From Multichannel EEG-MEG Signals

“…To quadratic SVM (QSVM) [30], k-nearest neighbor (KNN) [31], RF [32], LDA [33], and Naive Bayes (NB) [34] using performance metrics such as accuracy (Acc), sensitivity (Sen), specificity (Spe), and Fisher (F)1-score. These classifiers have been chosen based on their proven capability to investigate features with appropriate parameters and counteract over-fitting problems with less computational complexity in VSA decoding applications [3], [4], [18]. To prove its potential against the existing method, the proposed method has been tested for its mode-alignment property by aligning common frequency scales across multiple EEG-MEG data.…”

“…images from different categories are presented to the participants in experimental trails while their brain activities are recorded [1]. Researchers have explored several experimental techniques for recognizing visual objects, including EEG [2], MEG [3], near-infrared spectroscopy (NIRS) [4], electrocorticography (ECoG), and functional magnetic resonance imaging (fMRI) [1], [3]. Among these studies, MEG and EEG are the widely used techniques for visual recognition systems because it is non-invasive, cost-effective, and provide enhanced fine-grained analysis by identifying spatial, temporal, and spectral components underlying object category discrimination [1], [3].…”

“…Furthermore, the work presented by Zheng et al [7], in which the existing event-related potential (ERP)-long short-term memory (LSTM) framework was achieved an accuracy of 66.81% to recognise different stages of visual evoked potentials using EEG feature patterns across 72-channel. In recent works, [4] has introduced a visual object classification approach by adopting informative feature spaces on a single trail 128-channel EEG with supervised classifiers such as LDA, support vector machine (SVM), known as one versus rest (KOVR), random forest (RF), and XGBoost, and reported accuracy of 67.03%. Recently authors Cichy et al [1], [3] have extended the potential of the EEG and MEG multichannel data combinable to multivariate analysis of visual object problems.…”

“…This study demonstrated that the extracted RSA features from combined MEG and EEG signals enhance the corelation between temporal dynamics and category information in recognition of visual objects and deliver the highest classification performance than in MEG and EEG signals alone. However, in the above visual object classification approaches, the computed features from raw EEG or MEG multichannel signals show strong dependency on the formulation of the classification model, thus it may suffer in computing the most significant features with strong relevance to the visual object category [4]- [6], [8].…”

Domain knowledge-assisted multi-objective evolutionary algorithm for channel selection in brain-computer interface systems

“…Machine-learning-based methods (Siddiqui et al, 2020 ) generally select proper channels by training on the features extracted from the obtained EEG signals with the help of classic machine-learning techniques like the neural network. Wrapper-based methods (Liu Q. et al, 2021 ; Yavandhasani and Ghaderi, 2021 ) usually adopt predictors to solve the channel selection problem and tune wrappers according to the specific interaction between classifiers and datasets. As an efficient way to solve NP-compete problems, heuristic-searching-based methods have been adopted to solve channel selection problems successfully.…”