Characterizing Computer Access Using a One-Channel EEG Wireless Sensor

Molina, Alberto; Guerrero-Cubero, Jaime; Gómez, I.; Merino-Monge, Manuel; Silva-Silva, Juan I

doi:10.3390/s17071525

Cited by 11 publications

(13 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The sampling frequency was 256Hz. There are strong evidences from several studies which prove the efficiency of frontal EEG channel in studying attention-related tasks [11,14,22,24,38].…”

Section: Datamentioning

confidence: 99%

Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI

et al. 2019

View full text Add to dashboard Cite

Objective: Despite the effective application of deep learning in brain-computer interface (BCI) systems, the successful execution of this technique especially for inter-subject classification in cognitive BCI has not been accomplished yet. In this paper, we propose a framework based on deep convolutional neural network (CNN) to detect attentive mental state from single-channel raw electroencephalography (EEG) data. Approach: We develop an end-to-end deep CNN to decode the attentional information from EEG time-series. We also explore the consequence of input representations on the performance of deep CNN by feeding three different EEG representations into the network. To ensure the practical application of the proposed framework and avoid time-consuming re-trainings, we perform inter-subject transfer learning techniques as classification strategy. Eventually, to interpret the learned attentional patterns, we visualize and analyze the network perception of attention and nonattention classes. Main results: The average classification accuracy is 79.26% with only 15.83% of 120 subjects having the accuracy below 70% (a generally accepted threshold for BCI). This is while with inter-subject approach, it is literally hard to output high classification accuracy. This end-to-end classification framework surpasses the conventional classification methods for attention detection. The visualization results validate that the learned patterns from raw data are meaningful. Significance: This framework significantly improves the attention detection accuracy with inter-subject classification. Moreover, this study sheds light into the research on end-to-end learning; the proposed network is capable to learn from raw data with the least amount of pre-processing which in turn eliminates the extensive computational load of time-consuming data preparation and feature extraction.

show abstract

“…The sampling frequency was 256Hz. There are strong evidences from several studies which prove the efficiency of frontal EEG channel in studying attention-related tasks [11,14,22,24,38].…”

Section: Datamentioning

confidence: 99%

Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Besides brain electrical activity, EEG electrodes can record environmental electromagnetic interference or eye-blink artifacts [ 56 , 57 ]. Following the EEG preprocessing steps in [ 58 , 59 ], we applied normalization, artifacts removal and bandpass filtering on the extracted EEG signals.…”

Section: Methodsmentioning

confidence: 99%

Moment-to-Moment Continuous Attention Fluctuation Monitoring through Consumer-Grade EEG Device

Zhang

Yan

Sapkota

et al. 2021

Sensors

View full text Add to dashboard Cite

While numerous studies have explored using various sensing techniques to measure attention states, moment-to-moment attention fluctuation measurement is unavailable. To bridge this gap, we applied a novel paradigm in psychology, the gradual-onset continuous performance task (gradCPT), to collect the ground truth of attention states. GradCPT allows for the precise labeling of attention fluctuation on an 800 ms time scale. We then developed a new technique for measuring continuous attention fluctuation, based on a machine learning approach that uses the spectral properties of EEG signals as the main features. We demonstrated that, even using a consumer grade EEG device, the detection accuracy of moment-to-moment attention fluctuations was 73.49%. Next, we empirically validated our technique in a video learning scenario and found that our technique match with the classification obtained through thought probes, with an average F1 score of 0.77. Our results suggest the effectiveness of using gradCPT as a ground truth labeling method and the feasibility of using consumer-grade EEG devices for continuous attention fluctuation detection.

show abstract

“…It is not always easy for the experimenter to collect data and synchronize them with experimental phases, time intervals or simple events, when several devices and programs are running at the same time, as with multimodal interfaces [20]. To illustrate this, in [27], [26] two different systems were proposed to allow people with disabilities to access a computer. The former modulates the level of attention through an electroencephalography (EEG) sensor, whereas the latter requires residual movements detected by accelerometers or flex sensors.…”

Section: Introductionmentioning

confidence: 99%

An Easy-to-Use Multi-Source Recording and Synchronization Software for Experimental Trials

et al. 2020

Self Cite

View full text Add to dashboard Cite

A typical routine in many scientific studies consists of recording data from devices and identifying which segment of data corresponds with an experimental interval. However, many current applications have been designed to obtain and save data from one single device, and synchronizing data with the markers that delimit the test phases can be difficult. To address this issue, we have developed LSRec, which is based on Lab-Streaming Layer, a C++ library that allows data synchronization. LSRec is an easy-to-use, open-source, multi-platform, recording system developed on Java that can save data from several devices at the same time, while maintaining synchronization with the experimental phase markers. It supports three explicit sync methods: the first uses one-integer-channel input Lab-Streaming Layer streams. The others use TCP/UDP socket messages and allow any existing software that generates sync markers through TCP/UDP messages to be used under the same conditions. In LSRec, the markers are saved together with input data, facilitating the process of linking data with test stages. A prerequisite for recording software is to guarantee suitable timing performance, with no data loss and an easy user interface. These features were assessed for LSRec, with no data loss detected for 20 hours (25 sessions). We evaluated performance by measuring timestamp deviations for sync marker and input data. The sync methods exhibited an average of deviations of around −34µs (which is more than acceptable, e.g., in studies involving living beings), whereas the absolute value of deviation for one-channel data was lower than 40µs. Finally, we assessed the system's usability with the technology acceptance model 3 survey (6 volunteers). Subjects saw the software as useful, and intended to use it in the future. In conclusion, LSRec is a useful tool for those who need to record data from multiple sources, and maintain synchronization with experimental phases with low delay.

show abstract

Characterizing Computer Access Using a One-Channel EEG Wireless Sensor

Cited by 11 publications

References 66 publications

Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI

Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI

Moment-to-Moment Continuous Attention Fluctuation Monitoring through Consumer-Grade EEG Device

An Easy-to-Use Multi-Source Recording and Synchronization Software for Experimental Trials

Contact Info

Product

Resources

About