Arrangements of Resting State Electroencephalography as the Input to Convolutional Neural Network for Biometric Identification

Lai, Chi Qin; Ibrahim, Haidi; Abdullah, Mohd Zaid; Abdullah, Jafri Malin; Suandi, Shahrel Azmin; Azman, Azlinda

doi:10.1155/2019/7895924

Cited by 17 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Physionet data set used by the ref. [4][5][6], [11], [17], [20], [23], [25], [27][28], [33], and [37][38][39] and work on the publicly available dataset consisting of EEG of 109 participants completing various motor/imagery duties, it's a popular benchmark for biometric with EEG. In [9] datasets from four different experiments measuring endogenous brain functions (driving fatigue and emotion) in addition to time-locked artificially created brain responses from 157 subjects, [5] datasets including emotion and combined data.…”

Section: Datasets and Devicesmentioning

confidence: 99%

“…Additionally, the signals recorded by EEG systems must be carefully analyzed and interpreted to get useful data regarding the brain's electrical activity. In [2], [4][5][6], [9], [11], [14], [16][17][18][19], [20], [23], [25], [27][28], [30], [33], and [37][38][39] worked on BCI2000 system to record and analyzed EEG using 32 electrodes, while [1][8] [34][40] used AgCl electrodes EEG signals were recorded using a (Bio semi) Active Two system, EEG data were collected at a 512 sampling rate (Hz), AgCl with 32 electrodes works on the (10-20) of the international systems. Another device was using named GALILEO BE Light amplifier equipped with 19 channels/electrodes.…”

Section: Datasets and Devicesmentioning

confidence: 99%

“…The design was tested on two baseline scenarios in (RS), (EC, and EO) as shown in [4], [14], [20], [25], [28], [36][37][38][39]. in [10] EC Use this trigger only with the data to build the model and extract the results.…”

Section: Transcranial Magnetic Stimulation (Tms)mentioning

confidence: 99%

See 2 more Smart Citations

Electroencephalographic (EEG) based Deep Learning (DL): A Comparative Review

2023

IJSER

View full text Add to dashboard Cite

Deep learning (DL) has recently shown great promise in supporting knowledge of electroencephalographic (EEG) as a result of its ability to discover visual features (feature representation) from original (raw) data. This review will look at the latest developments in the research area of the EEG by analyzing a largest amount of the recent and definitive publications on EEG based on DL for biometrics identification. It covers the latest developments in different parts of the DL-EEG methodology and offers valuable information about them in order to improve its implementation. Also, it will provide interested researchers with a brief overview of the prospects of applying DL typical EEG processing methods. In addition to highlighting interesting methods and trends that used to acquisition and analyses brain signals, the stimulations, feature extractions and classifications. We summarize our review in some recommendations and proposals in the hope of promoting effective viable research in this field. We have highlighted interesting approaches and directions from this extensive research in order to provide ample information for future research. This review revealed that the duration of time spent trying to collect EEG data ranged from (10) minutes or less to a long time of hours, and interestingly, we found that more than 50 percent of the research design their models using publicly available datasets. Furthermore, There about half of the researchers used unprocessed or preprocessed EEG samples to train their models. Compared to traditional approaches, DL had an improvement in accuracy of 4% among the most applicable studies. More importantly, we discovered that the majority of previous studies have poor reproducibility: it is extremely difficult or impossible to replicate the majority of research due to a lack of data and codes. The importance of the paper lies in helping the research community to share and develop work more effectively. And We'll also offer a list of suggestions for more studies in the future.

show abstract

Section: Datasets and Devicesmentioning

confidence: 99%

Section: Datasets and Devicesmentioning

confidence: 99%

See 1 more Smart Citation

Electroencephalographic (EEG) based Deep Learning (DL): A Comparative Review

2023

IJSER

View full text Add to dashboard Cite

show abstract

“…The EEG signals were obtained using 64 electrodes to record brain signals from 64 different locations on the scalp. All electrodes were placed following an international standard of 10-10 electrode configuration [16,43]. The CP z (i.e., equivalent to Ch-32) was set as an EOG channel for tracking the eye movement and blinking artifacts.…”

Section: Eeg Recordingsmentioning

confidence: 99%

Prediction of Recovery from Traumatic Brain Injury with EEG Power Spectrum in Combination of Independent Component Analysis and RUSBoost Model

et al. 2022

Self Cite

View full text Add to dashboard Cite

The computational electroencephalogram (EEG) is recently garnering significant attention in examining whether the quantitative EEG (qEEG) features can be used as new predictors for the prediction of recovery in moderate traumatic brain injury (TBI). However, the brain’s recorded electrical activity has always been contaminated with artifacts, which in turn further impede the subsequent processing steps. As a result, it is crucial to devise a strategy for meticulously flagging and extracting clean EEG data to retrieve high-quality discriminative features for successful model development. This work proposed the use of multiple artifact rejection algorithms (MARA), which is an independent component analysis (ICA)-based algorithm, to eliminate artifacts automatically, and explored their effects on the predictive performance of the random undersampling boosting (RUSBoost) model. Continuous EEG were acquired using 64 electrodes from 27 moderate TBI patients at four weeks to one-year post-accident. The MARA incorporates an artifact removal stage based on ICA prior to RUSBoost, SVM, DT, and k-NN classification. The area under the curve (AUC) of RUSBoost was higher in absolute power spectral density (PSD) in AUCδ = 0.75, AUC α = 0.73 and AUCθ = 0.71 bands than SVM, DT, and k-NN. The MARA has provided a good generalization performance of the RUSBoost prediction model.

show abstract

“…Due to the nature of the EEG, resting-state EEG-based [ 22 , 28 , 29 , 30 ] verification with eyes open may find its place in a different application than other biometrics, like fingerprints, for example. It can be used in systems that require continuous identity confirmation, in brain-computer interfaces, or in EEG neurofeedback therapies based on participant-specific protocols where the subject must be verified to have the particular protocol applied.…”

Section: Introductionmentioning

confidence: 99%

Leveraging Multiple Distinct EEG Training Sessions for Improvement of Spectral-Based Biometric Verification Results

Plucińska

Jędrzejewski

Malinowska

et al. 2023

Sensors

View full text Add to dashboard Cite

Most studies on EEG-based biometry recognition report results based on signal databases, with a limited number of recorded EEG sessions using the same single EEG recording for both training and testing a proposed model. However, the EEG signal is highly vulnerable to interferences, electrode placement, and temporary conditions, which can lead to overestimated assessments of the considered methods. Our study examined how different numbers of distinct recording sessions used as training sessions would affect EEG-based verification. We analyzed the original data from 29 participants with 20 distinct recorded sessions each, as well as 23 additional impostors with only one session each. We applied raw coefficients of power spectral density estimate, and the coefficients of power spectral density estimate converted to the decibel scale, as the input to a shallow neural network. Our study showed that the variance introduced by multiple recording sessions affects sensitivity. We also showed that increasing the number of sessions above eight did not improve the results under our conditions. For 15 training sessions, the achieved accuracy was 96.7 ± 4.2%, and for eight training sessions and 12 test sessions, it was 94.9 ± 4.6%. For 15 training sessions, the rate of successful impostor attacks over all attack attempts was 3.1 ± 2.2%, but this number was not significantly different from using six recording sessions for training. Our findings indicate the need to include data from multiple recording sessions in EEG-based recognition for training, and that increasing the number of test sessions did not significantly affect the obtained results. Although the presented results are for the resting-state, they may serve as a baseline for other paradigms.

show abstract

Arrangements of Resting State Electroencephalography as the Input to Convolutional Neural Network for Biometric Identification

Cited by 17 publications

References 29 publications

Electroencephalographic (EEG) based Deep Learning (DL): A Comparative Review

Electroencephalographic (EEG) based Deep Learning (DL): A Comparative Review

Prediction of Recovery from Traumatic Brain Injury with EEG Power Spectrum in Combination of Independent Component Analysis and RUSBoost Model

Leveraging Multiple Distinct EEG Training Sessions for Improvement of Spectral-Based Biometric Verification Results

Contact Info

Product

Resources

About