Electroencephalography based imagined alphabets classification using spatial and time‐domain features

Agarwal, Prabhakar; Kumar, Sandeep

doi:10.1002/ima.22655

Cited by 11 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In one of our previous works [20] on the SI data [28], the imagined vowels/a/and/u/were classified with a maximum average accuracy of 70% in the alpha band. In Agarwal and Kumar [36], an attempt to provide the frequency band for each individual imagined English alphabet in which it can be classified with the highest accuracy had been done. The results were inclined toward the alpha and theta bands.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Electroencephalography‐based imagined speech recognition using deep long short‐term memory network

Agarwal

Kumar

2022

ETRI Journal

Self Cite

View full text Add to dashboard Cite

This article proposes a subject-independent application of brain-computer interfacing (BCI). A 32-channel Electroencephalography (EEG) device is used to measure imagined speech (SI) of four words (sos, stop, medicine, washroom) and one phrase (come-here) across 13 subjects. A deep long short-term memory (LSTM) network has been adopted to recognize the above signals in seven EEG frequency bands individually in nine major regions of the brain.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The procedure of the experiment is shown in Figure 1. The authors have adopted a similar experimental paradigm in decoding imagined English alphabets [36]. The signals of the imagined words were recorded according to international 10–20 system placement of electrodes [37] by a 32 channel EEG device ‘Mobita’ produced by TMSi systems.…”

Section: Methodsmentioning

confidence: 99%

Electroencephalography‐based imagined speech recognition using deep long short‐term memory network

Agarwal

Kumar

2022

ETRI Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…The first level DWT decomposes a signal into approximation coefficients (AC 1 ) and detail coefficients (DC 1 ), and further, at each subsequent level, the approximation coefficients (AC n ) are further divided into detail coefficients (DC n+1 ) and approximation coefficients (AC n+1 ). The order 4 Daubechies (db4) wavelet was used in the experiment because its smoothing property made it more suited to detecting changes in EEG signals [39] and also because research has suggested that the db4 wavelet basis is useful in imagined speech [18,19,40]. The DWT was employed at level 7 of signal decomposition using the db4 mother wavelet.…”

Section: Eeg Signal Processingmentioning

confidence: 99%

“…Varshney and Khan [18] experimented with the imagined speech EEG signals of six words using the discrete wavelet transform (DWT) to extract features and classify the signals using different models (RF and SVM) and obtained an average classification accuracy of 28.61%. Agarwal and Kumar [19] studied the use of DWT for denoise and CSP signal processing techniques to extract spatial characteristics from imagined speech of the twenty-six English alphabet EEG signals and classify the signals employing RF and SVM, obtaining a classification accuracy of 77.97%. Einizade et al [20], in the experiment decoding imagined speech of three words, exploited graph signal processing and graph learning-based signal characteristics of EEG signals to boost the effectiveness of SVM-based classifiers and achieved an accuracy of 50.10%.…”

Section: Introductionmentioning

confidence: 99%

Decoding of imagined speech electroencephalography neural signals using transfer learning method

Mahapatra,

Bhuyan

2023

J. Phys. Commun.

View full text Add to dashboard Cite

The use of brain-computer interfaces to produce imagined speech from brain waves has the potential to assist individuals with difficulty producing speech or communicating silently. The decoding of covert speech has been observed to have limited efficacy due to the diverse nature of the associated measured brain waves and the limited number of covert speech databases. As a result, traditional machine learning algorithms for learning and inference are challenging, and one of the real alternatives could be to leverage transfer of learning. The main goals of this research were to create a new deep learning (DL) framework for decoding imagined speech electroencephalography (EEG) signals tasks using transfer learning and to transfer the model learning of the source task of an imagined speech EEG dataset to the model training on the target task of another imagined speech EEG dataset, essentially the cross-task learning transfer of discriminative characteristics of the source task to the target task of imagined speech. The experiment was carried out using two distinct open-access EEG datasets, FEIS and KaraOne, that recorded the imagined speech classes of neural signals from multiple individuals. The target FEIS model and the target KaraOne model for multiclass classification exhibit overall accuracy of 80.65% and 81.32%, respectively, according to the proposed transfer learning. The experiment results indicate that the cross-task deep transfer learning design reliably classifies the imagined speech EEG signals by applying the source task learning to the target task learning. The findings suggest the feasibility of a consistent strategy for classifying multiclass imagined speech with transfer learning, which could thereby open up the possibility of future investigation into cross-task imagined speech classification knowledge usability for generalization of new imagined speech prompts.

show abstract

“…The decomposition was computed by repeatedly filtering the discrete signals up to a predefined level. Earlier studies have demonstrated the effectiveness of utilizing Daubechies-4 (order 4) wavelet feature extraction for the classification of different types of EEG signals [38][39][40][41]. In this study, we applied the DWT using the Daubechies-4 wavelet with two-level decomposition on EPOC and the three-level decomposition on MUSE for denoising and information extraction.…”

Section: Eeg Signal Preprocessingmentioning

confidence: 99%

EEG-based classification of imagined digits using a recurrent neural network

Mahapatra

Bhuyan

2023

J. Neural Eng.

View full text Add to dashboard Cite

Objective. In recent years, imagined speech brain-computer (machine) interface applications have been an important field of study that can improve the lives of patients with speech problems through alternative verbal communication. This study aimed to classify the imagined speech of numerical digits from electroencephalography (EEG) signals by exploiting the past and future temporal characteristics of the signal using several deep learning models. Approach. This study proposed a methodological combination of EEG signal processing techniques and deep learning models for the recognition of imagined speech signals. The EEG signals were filtered and preprocessed using the discrete wavelet transformation (DWT) to remove artifacts and retrieve feature information. To classify the preprocessed imagined speech neural signals, multiple versions of multlayer bidirectional recurrent neural networks were used. Main results. The method was examined by leveraging MUSE and EPOC signals from MNIST imagined digits in the MindBigData open-access database. The presented methodology’s classification performance accuracy was noteworthy, with the model’s multiclass overall classification accuracy reaching a maximum of 96.18 percent on MUSE signals and 71.60 percent on EPOC signals. Significance. Furthermore, this research shows that the proposed signal preprocessing approach and the stacked bidirectional recurrent network model are suitable for extracting the high temporal resolution of EEG signals in order to classify imagined digits, indicating the unique neural identity of each imagined digit class that distinguished it from the others.

show abstract

Electroencephalography based imagined alphabets classification using spatial and time‐domain features

Cited by 11 publications

References 45 publications

Electroencephalography‐based imagined speech recognition using deep long short‐term memory network

Electroencephalography‐based imagined speech recognition using deep long short‐term memory network

Decoding of imagined speech electroencephalography neural signals using transfer learning method

EEG-based classification of imagined digits using a recurrent neural network

Contact Info

Product

Resources

About