Objective: Blink-related features derived from electroencephalography (EEG) have recently arisen as a meaningful measure of drivers cognitive state. Combined with band power features of low-channel prefrontal EEG data, blink-derived features enhance the detection of driver drowsiness. Yet, it remains unanswered whether synergy of combined blink and EEG band power features for the detection of driver drowsiness may be further boosted if a proper eye blink removal is also applied before EEG analysis. This paper proposes an algorithm for simultaneous eye blink feature extraction and elimination from low-channel prefrontal EEG data. Methods: Firstly, eye blink intervals (EBIs) are identified from the Fp1 EEG channel using variational mode extraction, and then blink-related features are derived. Secondly, the identified EBIs are projected to the rest of EEG channels and then filtered by a combination of principal component analysis and discrete wavelet transform. Thirdly, a support vector machine with 10-fold cross-validation is employed to classify alert and drowsy states from the derived blink and filtered EEG band power features. Main results: When compared the synergy of eye blink and EEG features before and after filtering by the proposed algorithm, a significant improvement in the mean accuracy of driver drowsiness detection was achieved (71.2% vs. 78.1%, p<0.05). Significance: This paper validates a novel view of eye blinks as both a source of information and artifacts in EEG-based driver drowsiness detection.
8 November 2014 in Pattern Analysis and Applications
The aim of this paper is to present a new algorithm for feature selection, called a genetic algorithm with aggressive mutation. The paper presents both the theoretic background of the algorithm and its application for feature selection in the brain–computer interface (BCI) domain. To fully present the potential of the algorithm and to verify its practical usability, it is compared with other methods commonly used in BCI research. The practical application of the proposed algorithm is presented via a benchmark set submitted to the second BCI Competition (data set III—motor imagery).
Objective. The paper aims to present a method that enables the application of independent component analysis (ICA) to a low-channel EEG recording. The idea behind the method (called moving average ICA or MAICA) is to extend the original low-sensor matrix of signals by applying a set of zero-phase moving average filters to each of the recorded signals.Approach. The paper discusses the theoretical background of the MAICA algorithm and verifies its usefulness under three exemplary settings: (i) a pure mathematic system composed of ten source sinusoids; (ii) real EEG data recorded from 64 channels; (iii) real EEG data recorded from five subjects during 200 trials with motor imagery brain–computer interface.Main results. The first system shows that MAICA is able to decompose two mixed signals (composed of ten source sinusoids) into ten components with an extremely high correlation between the source patterns and identified components (99%–100%). The second system shows that when used over five channels, MAICA is able to recognize more artefact components than those recognized by classic ICA used over 64 channels. Finally, the third system demonstrates that MAICA is capable of working in an online mode without significant delays; the additional time needed to run MAICA for one trial was less than 6ms in the survey reported in the paper.Significance. The method presented in the paper should have a significant impact on all areas of medical signal processing where a large number of known and/or unknown patterns have to be retrieved in real time from complex signals recorded from a small number of external/internal body sensors.
As a result of the increasing role of online advertising and strong competition among advertisers, intrusive techniques are commonly used to attract web users’ attention. Moreover, since marketing content is usually delivered to the target audience when they are performing typical online tasks, like searching for information or reading online content, its delivery interrupts the web user’s current cognitive process. The question posed by many researchers in the field of online advertising is: how should we measure the influence of interruption of cognitive processes on human behavior and emotional state? Much research has been conducted in this field; however, most of this research has focused on monitoring activity in the simulated environment, or processing declarative responses given by users in prepared questionnaires. In this paper, a more direct real-time approach is taken, and the effect of the interruption on a web user is analyzed directly by studying the activity of his brain. This paper presents the results of an experiment that was conducted to find the brain activity patterns associated with interruptions of the cognitive process by showing internet advertisements during a text-reading task. Three specific aspects were addressed in the experiment: individual patterns, the consistency of these patterns across trials, and the intra-subject correlation of the individual patterns. Two main effects were observed for most subjects: a drop in activity in the frontal and prefrontal cortical areas across all frequency bands, and significant changes in the frontal/prefrontal asymmetry index.
This study presents the results of an experiment designed to investigate whether marketing videos containing mixed emotional content can sustain consumers interest longer compared to videos conveying a consistent emotional message. During the experiment, thirteen participants, wearing EEG (electroencephalographic) caps, were exposed to eight marketing videos with diverse emotional tones. Participant engagement was measured with an engagement index, a metric derived from the power of brain activity recorded over the frontal and parietal cortex and computed within three distinct frequency bands: theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). The outcomes indicated a statistically significant influence of emotional content type (mixed vs. consistent) on the duration of user engagement. Videos containing a mixed emotional message were notably more effective in sustaining user engagement, whereas the engagement level for videos with a consistent emotional message declined over time. The principal inference drawn from the study is that advertising materials conveying a consistent emotional message should be notably briefer than those featuring a mixed emotional message to achieve an equivalent level of message effectiveness, measured through engagement duration.