Optimized Bi-Objective EEG Channel Selection and Cross-Subject Generalization With Brain–Computer Interfaces

Handiru, Vikram Shenoy; Prasad, Vinod A.

doi:10.1109/thms.2016.2573827

Cited by 80 publications

(45 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another advantage of using EAs as the search algorithm within the wrapper procedure is that they are well-suited to solve MOOPs, allowing the search of solutions that, having a high training accuracy, meet other objectives too, such as a small number of selected features, a high generalization capability, or even anatomical and functional relevance of EEG channels [53]. Although there are currently several well known implementations of the MOEA concept, such as MOEA/D [54], PAES [55], SPEA2 [56] or NSGA-II [44], since the aim of this paper is not to analyze the characteristics of the a specific MOEA but rather applying a MOEA as the search algorithm for the proposed wrapper method, NSGA-II has been used because it is quite well known and widely used.…”

Section: Proposed Multi-objective Evolutionary Wrapper Methodsmentioning

confidence: 99%

A new multi-objective wrapper method for feature selection – Accuracy and stability analysis for BCI

et al. 2019

View full text Add to dashboard Cite

Feature selection is an important step in building classifiers for high-dimensional data problems, such as EEG classification for BCI applications. This paper proposes a new wrapper method for feature selection, based on a multi-objective evolutionary algorithm, where the representation of the individuals or potential solutions, along with the breeding operators and objective functions, have been carefully designed to select a small subset of features that has good generalization capability, trying to avoid the over-fitting problems that wrapper methods usually suffer. A novel feature ranking procedure is also proposed in order to analyze the stability of the proposed wrapper method. Four different classification schemes have been applied within the proposed wrapper method in order to evaluate its accuracy and stability for feature selection on a real motor imagery dataset. Experimental results show that the wrapper method presented in this paper is able to obtain very small subsets of features, which are quite stable and also achieve high classification accuracy, regardless of the classifiers used.

show abstract

Section: Proposed Multi-objective Evolutionary Wrapper Methodsmentioning

confidence: 99%

A new multi-objective wrapper method for feature selection – Accuracy and stability analysis for BCI

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Most previous BCI studies used relatively small to moderate training sample size (~8-30 subjects) and used all subjects' training data to train one model (training set usually divided into 80% of each user's EEG data and the remaining was held out for testing). Then, the accuracy of this trained model has been reported on each user's test data separately [28], [41], [45], [64], [66], [71], [72]. The motivation in the previous studies to train a single model on all users' training data, while the model was tested on each individual separately was likely, because the classification performance of most classifiers usually increase with more training data.…”

Section: Introductionmentioning

confidence: 99%

“…Several studies [1], [7], [28], [32], [40], [42], [44], [45], [50], [51], [62], [66], [71], [72], [77] have investigated the classification of MI-EEG to develop a BCI system that can provide feedback during MI training and eventually use the BCI to enhance the life of patients with disabilities and paralysis.…”

Section: Introductionmentioning

confidence: 99%

Motor Imagery Classification for Brain Computer Interface Using Deep Metric Learning

2020

View full text Add to dashboard Cite

Deep metric learning (DML) has achieved state-of-the-art results in several deep learning applications. However, this type of deep learning models has not been tested on the classification of electrical brain waves (EEG) for brain computer interface (BCI) applications. For the first time, we propose a triplet network to classify motor imagery (MI) EEG signals. Stockwell Transform has been used for converting the EEG signals in the time domain into the frequency domain, which resulted in improved DML classification accuracy in comparison to DML with Short Term Fourier Transform (0.647 vs. 0.431). DML model was trained with a topogram of concatenated 64 EEG channel spectrograms. The training batch was comprised of triplet pairs of the anchor, positive, and negative labeled epochs. The triplet network was able to train an embedding feature space that minimized the Euclidean distance between the embeddings of spectrograms of the same class and increased the distance between the embeddings of different labeled images. The proposed method has been tested on an EEG dataset of 109 untrained subjects. We showed that the DML classifier is able to converge with an extremely small number of training samples (~120 EEG trials) for only one subject per model, mitigating the well-known issue of the large inter-individual variability of human MI-BCI EEG which degrades the classification performance. The proposed preprocessing pipeline and the Triplet Network provide a promising method to classify MI-BCI EEG signals with much less training samples than the previous methods.

show abstract

“…Therefore, it is vital to opt for an effective solution to select the optimal number of channels rather than using all channels for processing and classification. Some researchers have used feature selection algorithms after applying the channel selection algorithms to further improve the system performance [22]. Subset channels are selected based on certain criteria that usually incorporates all aspects including channel location, dependency and redundancy [23].…”

Section: Introductionmentioning

confidence: 99%

Filtering techniques for channel selection in motor imagery EEG applications: a survey

2019

View full text Add to dashboard Cite

Brain Computer Interface (BCI) Systems are used in a wide range of applications such as communication, neuro-prosthetic and environmental control for disabled persons using robots and manipulators. A typical BCI system uses different types of inputs; however, Electroencephalography (EEG) signals are most widely used due to their non-invasive EEG electrodes, portability, and cost efficiency. The signals generated by the brain while performing or imagining a motor related task (Motor Imagery (MI)) signals are one of the important inputs for BCI applications. EEG data is usually recorded from more than 100 locations across the brain, so efficient channel selection algorithms are of great importance to identify optimal channels related to a particular application. The main purpose of applying channel selection is to reduce computational complexity while analysing EEG signals, improve classification accuracy by reducing over-fitting, and decrease setup time. Different channel selection evaluation algorithms such as filtering, wrapper, and hybrid methods have been used for extracting optimal channel subsets by using predefined criteria. After extensively reviewing the literature in the field of EEG channel

show abstract

Optimized Bi-Objective EEG Channel Selection and Cross-Subject Generalization With Brain–Computer Interfaces

Cited by 80 publications

References 39 publications

A new multi-objective wrapper method for feature selection – Accuracy and stability analysis for BCI

A new multi-objective wrapper method for feature selection – Accuracy and stability analysis for BCI

Motor Imagery Classification for Brain Computer Interface Using Deep Metric Learning

Filtering techniques for channel selection in motor imagery EEG applications: a survey

Contact Info

Product

Resources

About