Detection of lapses in responsiveness from the EEG

Designing an Embedded Feature Selection Algorithm for a Drowsiness Detector Model Based on Electroencephalogram Data

Driver fatigue reduces the safety of traditional driving and limits the widespread adoption of self-driving cars; hence, the monitoring and early detection of drivers' drowsiness plays a key role in driving automation. When representing the drowsiness indicators as large feature vectors, fitting a machine learning model to the problem becomes challenging, and the problem's perspicuity decreases, making dimensionality reduction crucial in practice. For this reason, we propose an embedded feature selection algorithm that can be later utilized as a building block in the system development of a neural network-based drowsiness detector. We have adopted a technique: a so-called Feature Prune Layer is placed in front of the first layer in the architecture; as a result, its weights change regarding the importance of the corresponding input features and are deleted iteratively until the desired number is reached. We test the algorithm on EEG data, as it is one of the best indicators of drowsiness based on the literature. The proposed FS algorithm is able to reduce the original feature set by 95% with only 1% degradation in precision, while the precision increases by 1.5% and 2.7% respectively when selecting the top 10% and top 20% of the initial features. Moreover, the proposed method outperforms the widely popular Principal Component Analysis and the Chi-squared test when reducing the original feature set by 95%: it achieves 24.3% and 3.2% higher precision respectively.

Effects of total sleep deprivation on execution lapses during vigilance tasks

Total sleep deprivation (TSD) results in reduced efficiency of cognitive resources. Moreover, when the available cognitive resources are less than required, individuals exhibit lapses in responsiveness. Accordingly, this study explored the effects of TSD on executive function and the characteristics of execution lapses. Functional near-infrared spectroscopy was used to monitor the prefrontal cortex's functional connections in resting and tasking states for various sleep deprivation durations. Data from participants' attentional performance test and self-reported fatigue were collected over 30 hours of wakefulness. Task performance was compared based on time of day, time on task, and reaction time. The results show that participants' arousal level significantly decreased post 14 hours (P < .05), while sleepiness increased. The prefrontal cortex connection and attentional performance dropped at the Window of Circadian Low (3:00 ~ 6:00). The number of execution lapses was higher during the initiation, inhibition, and fatigue phases and rose markedly post 14 hours of wakefulness. We conclude that maintaining better inhibition control requires a reasonable extension of the reaction time. Moreover, subjective perception is significantly correlated with task performance and right prefrontal connection strength. This study presents the scientific evidence for measures to address consistently long working hours and disrupted circadian rhythms.

Classification of alcoholic EEG signals using wavelet scattering transform-based features

Following the research question and the relevant dataset, feature extraction is the most important component of machine learning and data science pipelines. The wavelet scattering transform (WST) is a recently developed knowledge-based feature extraction technique and is structurally like a convolutional neural network (CNN). It preserves information in high-frequency, is insensitive to signal deformations, and generates low variance features of real-valued signals generally required in classification tasks. With data from a publicly-available UCI database, we investigated the ability of WST-based features extracted from multichannel electroencephalogram (EEG) signals to discriminate 1.0-s EEG records of 20 male subjects with alcoholism and 20 male healthy subjects. Using record-wise 10-fold cross-validation, we found that WST-based features, inputted to a support vector machine (SVM) classifier, were able to correctly classify all alcoholic and normal EEG records. Similar performances were achieved with 1D CNN. In contrast, the highest independent-subject-wise mean 10-fold cross-validation performance was achieved with WST-based features fed to a linear discriminant (LDA) classifier. The results achieved with two 10-fold cross-validation approaches suggest that the WST together with a conventional classifier is an alternative to CNN for classification of alcoholic and normal EEGs. WST-based features from occipital and parietal regions were the most informative at discriminating between alcoholic and normal EEG records.

Microsleep versus Sleep Onset Latency during Maintenance Wakefulness Tests: Which One Is the Best Marker of Sleepiness?

The interpretation of the Maintenance Wakefulness Test (MWT) relies on sleep onset detection. However, microsleeps (MSs), i.e., brief periods of sleep intrusion during wakefulness, may occur before sleep onset. We assessed the prevalence of MSs during the MWT and their contribution to the diagnosis of residual sleepiness in patients treated for obstructive sleep apnea (OSA) or hypersomnia. The MWT of 98 patients (89 OSA, 82.6% male) were analyzed for MS scoring. Polysomnography parameters and clinical data were collected. The diagnostic value for detecting sleepiness (Epworth Sleepiness Scale > 10) of sleep onset latency (SOL) and of the first MS latency (MSL) was assessed by the area under the receiver operating characteristic (ROC) curve (AUC, 95% CI). At least one MS was observed in 62.2% of patients. MSL was positively correlated with SOL (r = 0.72, p < 0.0001) but not with subjective scales, clinical variables, or polysomnography parameters. The use of SOL or MSL did not influence the diagnostic performance of the MWT for subjective sleepiness assessment (AUC = 0.66 95% CI (0.56, 0.77) versus 0.63 95% CI (0.51, 0.74)). MSs are frequent during MWTs performed in patients treated for sleep disorders, even in the absence of subjective sleepiness, and may represent physiological markers of the wake-to-sleep transition.

Reservoir computing approaches to microsleep detection

The detection of microsleeps in a wide range of professionals working in high-risk occupations is very important to workplace safety. A microsleep classifier is presented that employs a reservoir computing (RC) methodology. Specifically, echo state networks (ESN) are used to enhance previous benchmark performances on microsleep detection. A clustered design using a novel ESN-based leaky integrator is presented. The effectiveness of this design lies with the simplicity of using a fine-grained architecture, containing up to 8 neurons per cluster, to capture individualized state dynamics and achieve optimal performance. This is the first study to have implemented and evaluated EEG-based microsleep detection using RC models for the detection of microsleeps from the EEG. Microsleep state detection was achieved using a cascaded ESN classifier with leaky-integrator neurons employing 60 principal components from 544 power spectral features. This resulted in a leave-one-subject-out average detection in performance of Φ= 0.51 ± 0.07 (mean ± SE), AUC-ROC = 0.88 ± 0.03, and AUC-PR = 0.44 ± 0.09. Although performance of EEG-based microsleep detection systems is still considered modest, this refined method achieved a new benchmark in microsleep detection.

Decoding Analysis of Alpha Oscillation Networks on Maintaining Driver Alertness

The countermeasure of driver fatigue is valuable for reducing the risk of accidents caused by vigilance failure during prolonged driving. Listening to the radio (RADIO) has been proven to be a relatively effective “in-car” countermeasure. However, the connectivity analysis, which can be used to investigate its alerting effect, is subject to the issue of signal mixing. In this study, we propose a novel framework based on clustering and entropy to improve the performance of the connectivity analysis to reveal the effect of RADIO to maintain driver alertness. Regardless of reducing signal mixing, we introduce clustering algorithm to classify the functional connections with their nodes into different categories to mine the effective information of the alerting effect. Differential entropy (DE) is employed to measure the information content in different brain regions after clustering. Compared with the Louvain-based community detection method, the proposed method shows more superior ability to present RADIO effectin confused functional connection matrices. Our experimental results reveal that the active connection clusters distinguished by the proposed method gradually move from frontal region to parieto-occipital regionwith the progress of fatigue, consistent with the alpha energy changes in these two brain areas. The active class of the clusters in parieto-occipital region significantly decreases and the most active clusters remain in the frontal region when RADIO is taken. The estimation results of DE confirm the significant change (p < 0.05) of information content due to the cluster movements. Hence, preventing the movement of the active clusters from frontal region to parieto-occipital region may correlate with maintaining driver alertness. The revelation of alerting effect is helpful for the targeted upgrade of fatigue countermeasures.

Automatic detection of microsleep episodes with feature-based machine learning

STUDY OBJECTIVES

Microsleep episodes (MSEs) are brief episodes of sleep, mostly defined to be shorter than 15 s. In the electroencephalogram (EEG), MSEs are mainly characterized by a slowing in frequency. The identification of early signs of sleepiness and sleep (e.g. MSEs) is of considerable clinical and practical relevance. Under laboratory conditions, the maintenance of wakefulness test (MWT) is often used for assessing vigilance.

METHODS

We analyzed MWT recordings of 76 patients referred to the Sleep-Wake-Epilepsy-Center. MSEs were scored by experts defined by the occurrence of theta dominance on ≥1 occipital derivation lasting 1-15 s, whereas the eyes were at least 80% closed. We calculated spectrograms using an autoregressive model of order 16 of 1 s epochs moved in 200 ms steps in order to visualize oscillatory activity and derived seven features per derivation: power in delta, theta, alpha and beta bands, ratio theta/(alpha + beta), quantified eye movements, and median frequency. Three algorithms were used for MSE classification: support vector machine (SVM), random forest (RF), and an artificial neural network (long short-term memory [LSTM] network). Data of 53 patients were used for the training of the classifiers, and 23 for testing.

RESULTS

MSEs were identified with a high performance (sensitivity, specificity, precision, accuracy, and Cohen's kappa coefficient). Training revealed that delta power and the ratio theta/(alpha + beta) were most relevant features for the RF classifier and eye movements for the LSTM network.

CONCLUSIONS

The automatic detection of MSEs was successful for our EEG-based definition of MSEs, with good performance of all algorithms applied.

Neural Correlates of Attention Lapses During Continuous Tasks

Attention lapses (ALs) are common phenomenon, which can affect our performance and productivity by slowing or suspending responsiveness. Occurrence of ALs during continuous monitoring tasks, such as driving or operating machinery, can lead to injuries and fatalities. However, we have limited understanding of what happens in the brain when ALs intrude during such continuous tasks. Here, we analyzed fMRI data from a study, in which participants performed a continuous visuomotor tracking task during fMRI scanning. A total of 68 ALs were identified from 20 individuals, using visual rating of tracking performance and video-based eye-closure. ALs were found to be associated with increased BOLD fMRI activity partially in the executive control network, and sensorimotor network. Surprisingly, we found no evidence of deactivations.

Optimizing Computer-Brain Interface Parameters for Non-invasive Brain-to-Brain Interface

A non-invasive, brain-to-brain interface (BBI) requires precision neuromodulation and high temporal resolution as well as portability to increase accessibility. A BBI is a combination of the brain-computer interface (BCI) and the computer-brain interface (CBI). The optimization of BCI parameters has been extensively researched, but CBI has not. Parameters taken from the BCI and CBI literature were used to simulate a two-class medical monitoring BBI system under a wide range of conditions. BBI function was assessed using the information transfer rate (ITR), measured in bits per trial and bits per minute. The BBI ITR was a function of classifier accuracy, window update rate, system latency, stimulation failure rate (SFR), and timeout threshold. The BCI parameters, including window length, update rate, and classifier accuracy, were kept constant to investigate the effects of varying the CBI parameters, including system latency, SFR, and timeout threshold. Based on passively monitoring BCI parameters, a base ITR of 1 bit/trial was used. The optimal latency was found to be 100 ms or less, with a threshold no more than twice its value. With the optimal latency and timeout parameters, the system was able to maintain near-maximum efficiency, even with a 25% SFR. When the CBI and BCI parameters are compared, the CBI's system latency and timeout threshold should be reflected in the BCI's update rate. This would maximize the number of trials, even at a high SFR. These findings suggested that a higher number of trials per minute optimizes the ITR of a non-invasive BBI. The delays innate to each BCI protocol and CBI stimulation method must also be accounted for. The high latencies in each are the primary constraints of non-invasive BBI for the foreseeable future.

Detection and Prediction of Microsleeps from EEG using Spatio-Temporal Patterns

A microsleep is a brief lapse in performance due to an involuntary sleep-related loss of consciousness. These episodes are of particular importance in occupations requiring extended unimpaired visuomotor performance, such as driving. Detection and even prediction of microsleeps has the potential to prevent catastrophic events and fatal accidents. In this study, we examined detection and prediction of microsleeps using EEG data of 8 subjects who performed two 1-h sessions of continuous 1-D tracking. A regularized spatio-temporal filtering and classification (RSTFC) method was used to extract features from 5-s EEG segments. These features were then used to train three different linear classifiers: linear discriminant analysis (LDA), sparse Bayesian learning (SBL), and variational Bayesian logistic regression (VBLR). The performance of microsleep state detection and prediction was evaluated using leave-one-subject-out cross-validation. The detection performance measures were AUC_ROC 0.96, AUC_PR 0.52, and phi 0.47. As expected, prediction of microsleep states with a 0.25-s ahead prediction time resulted in slightly lower performances compared to the detection. Prediction performance measures were substantially higher than those achieved with log-power spectral features, i.e., AUC_ROC 0.95 (cf. 0.90), AUC_PR 0.50 (cf. 0.36), and phi 0.46 (cf. 0.34).

Deep Learning with Convolutional Neural Network for detecting microsleep states from EEG: A comparison between the oversampling technique and cost-based learning

Any occupation which involves critical decision making in real-time requires attention and concentration. When repetitive and expanded working periods are encountered, it can result in microsleeps. Microsleeps are complete lapses in which a subject involuntarily stops responding to the task that they are currently performing due to temporary interruptions in visual-motor and cognitive coordination. Microsleeps can last up to 15 s while performing a particular task. In this study, the ability of a convolutional neural network (CNN) to detect microsleep states from 16-channel EEG data from 8 subjects, performing a 1D visuomotor was explored. The data were highly imbalanced. When averaged across 8 subjects there were 17 responsive states for every microsleep state. Two approaches were used to handle the CNN training with data imbalance - oversampling the minority class and cost-based learning. The EEG was analysed using a 4-s epoch with a step size of 0.25 s. Leave-one-subject-out cross-validation was used to evaluate the performance. The performance measures used for assessing the detection capability of the CNN were: sensitivity, precision, phi, geometric mean (GM), AUC_ROC, and AUC_PR. The performance measures obtained using the oversampling and cost-based learning methods were: AUC_ROC = 0.90/0.90, AUC_PR = 0.41/0.41 and a phi = 0.42/0.40, respectively. Although the performances were similar, the cost-based learning method had a considerably shorter training time than the oversampling method.

An iterative cross-subject negative-unlabeled learning algorithm for quantifying passive fatigue

OBJECTIVE

This paper proposes an iterative negative-unlabeled (NU) learning algorithm for cross-subject detection of passive fatigue from labelled alert (negative) and unlabeled driving EEG data.

APPROACH

Unlike other studies which used manual labeling of the fatigue state, the proposed algorithm (PA) first iteratively uses 29 subjects' alert data and unlabeled driving data to identify the most fatigued block of EEG data in each subject in a cross-subject manner. Subsequently, the PA computes subjects' driving fatigue score. Repeated measures correlations of the score to EEG band powers are then performed.

MAIN RESULTS

The PA yields an averaged accuracy of 93.77%  ±  8.15% across subjects in detecting fatigue, which is significantly better than the various baselines. The fatigue scores obtained are also significantly positively correlated with theta band power and negatively correlated with beta band power that are known to respectively increase and decrease in presence of passive fatigue. There is a strong negative correlation with alpha band power as well.

SIGNIFICANCE

The proposed iterative NU learning algorithm is capable of labelling and quantifying the most fatigued block in a cross-subject manner despite the lack of ground truth in the fatigue levels of unlabeled driving EEG data. Together with the significant correlations with theta, alpha and beta band power, the results show promise in the application of the proposed algorithm to detect fatigue from EEG.

Ensemble learning based on overlapping clusters of subjects to predict microsleep states from EEG

Microsleeps are brief and involuntary instances of complete loss of sleep-related consciousness. We present a novel approach of creating overlapping clusters of subjects and training of an ensemble classifier to enhance the prediction of microsleep states from EEG. Overlapping clusters are created using Kullback-Leibler divergence between responsive state features of each pair of training subjects. Highly correlated features within each overlapping cluster are discarded. The remaining features are selected via Fisher score based ranking followed by an average of 5-fold cross-validation areas under the curves of receiver operating characteristics (AUCRoc) of a linear discriminant analysis (LDA) classifier. The decisions of LDA classifiers on overlapping clusters are fused using weighted average. We evaluated this new approach on 16-channel EEG data from 8 subjects who had performed a 1-D visuomotor task for two l-h sessions. Joint entropy features were extracted from a 5-s window of EEG with steps of 0.25 s Test performances were evaluated using leave-one-subject-out cross-validation. Our ensemble of overlapping clusters of subjects achieved a mean prediction performance, phi, of 0.42 compared with 0.39 for a single LDA classifier and 0.37 for generalized stacking.

Prediction of microsleeps using pairwise joint entropy and mutual information between EEG channels

Microsleeps are involuntary and brief instances of complete loss of responsiveness, typically of 0.5-15 s duration. They adversely affect performance in extended attention-driven jobs and can be fatal. Our aim was to predict microsleeps from 16 channel EEG signals. Two information theoretic concepts - pairwise joint entropy and mutual information - were independently used to continuously extract features from EEG signals. k-nearest neighbor (kNN) with k = 3 was used to calculate both joint entropy and mutual information. Highly correlated features were discarded and the rest were ranked using Fisher score followed by an average of 3-fold cross-validation area under the curve of the receiver operating characteristic (AUC_ROC). Leave-one-out method (LOOM) was performed to test the performance of microsleep prediction system on independent data. The best prediction for 0.25 s ahead was AUCROC, sensitivity, precision, geometric mean (GM), and φ of 0.93, 0.68, 0.33, 0.75, and 0.38 respectively with joint entropy using single linear discriminant analysis (LDA) classifier.

Temporal evolution of neural activity and connectivity during microsleeps when rested and following sleep restriction

Even when it is critical to stay awake, such as when driving, sleep deprivation weakens one's ability to do so by substantially increasing the propensity for microsleeps. Microsleeps are complete lapses of consciousness but, paradoxically, are associated with transient increases in cortical activity. But do microsleeps provide a benefit in terms of attenuating the need for sleep? And is the neural response to microsleeps altered by the degree of homeostatic drive to sleep? In this study, we continuously monitored eye-video, visuomotor responsiveness, and brain activity via fMRI in 20 healthy subjects during a 20-min visuomotor tracking task following a normally-rested night and a sleep-restricted (4-h) night. As expected, sleep restriction led to an increased number of microsleeps and an increased variability in tracking error. Microsleeps exhibited transient increases in regional activity in the fronto-parietal and parahippocampal area. Network analyses revealed divergent transient changes in the right fronto-parietal, dorsal-attention, default-mode, and thalamo-cortical functional networks. In all subjects, tracking error immediately following microsleeps was improved compared to before the microsleeps. Importantly, post-microsleep recovery in tracking response speed was associated with hyperactivation in the thalamo-cortical network. The temporal evolution of functional connectivity within the frontal and posterior nodes of the default-mode network and between the right fronto-parietal and default-mode networks was associated with temporal changes in visuomotor responsiveness. These findings demonstrate distinct brain-network-level changes in brain activity during microsleeps and suggest that neural activity in the thalamo-cortical network may facilitate the transient recovery from microsleeps. The temporal pattern of evolution in brain activity and performance is indicative of dynamic changes in vigilance during the struggle to stay awake following sleep loss.

Bayesian multi-subject factor analysis to predict microsleeps from EEG power spectral features

Prediction of an imminent microsleep has the potential to save lives and prevent catastrophic accidents. A microsleep is a brief episode of unintentional unconsciousness and, hence, loss of responsiveness. In this study, prediction of imminent microsleeps using EEG data from 8 subjects was examined. A novel Bayesian algorithm was proposed to identify common components of pre-microsleep activity in the EEG in all subjects and predict microsleeps 0.25 s ahead. To avoid overfitting, this model incorporates sparsity-promoting priors to automatically find the minimum number of components. Due to intractability of full Bayesian treatment, variational Bayesian was integrated to approximate posterior probabilities. To predict microsleeps, EEG log-power spectral features were extracted from a 5-s window. Bayesian multi-subject factor analysis was used to extract common microsleep patterns and transform all features into lower-dimension common-space features. Discrimination between responsive and microsleep instances was done with a single linear discriminant analysis (LDA) classifier. Performance of the proposed method was evaluated using leave-one-subject-out cross-validation. Our prediction system achieved moderate AUC_ROC and GM of 0.90 and 0.80, respectively, but with a relatively low precision of 0.29.

A software framework for real-time multi-modal detection of microsleeps

A software framework is described which was designed to process EEG, video of one eye, and head movement in real time, towards achieving early detection of microsleeps for prevention of fatal accidents, particularly in transport sectors. The framework is based around a pipeline structure with user-replaceable signal processing modules. This structure can encapsulate a wide variety of feature extraction and classification techniques and can be applied to detecting a variety of aspects of cognitive state. Users of the framework can implement signal processing plugins in C++ or Python. The framework also provides a graphical user interface and the ability to save and load data to and from arbitrary file formats. Two small studies are reported which demonstrate the capabilities of the framework in typical applications: monitoring eye closure and detecting simulated microsleeps. While specifically designed for microsleep detection/prediction, the software framework can be just as appropriately applied to (i) other measures of cognitive state and (ii) development of biomedical instruments for multi-modal real-time physiological monitoring and event detection in intensive care, anaesthesiology, cardiology, neurosurgery, etc. The software framework has been made freely available for researchers to use and modify under an open source licence.

Optimized echo state networks with leaky integrator neurons for EEG-based microsleep detection

The performance of a microsleep detection system was calculated in terms of its ability to detect the behavioural microsleep state (1-s epochs) from spectral features derived from 16-channel EEG sampled at 256 Hz. Best performance from a single classifier model was achieved using leaky integrator neurons on an echo state network (ESN) classifier with a mean phi correlation (φ) of 0.38 and accuracy of 67.3%. A single classifier model of ESN with sigmoidal inputs achieved φ of 0.20 and accuracy of 48.5% and a single classifier model of linear discriminant analysis (LDA) achieved φ of 0.31 and accuracy of 53.6%. However, combining the output of several single classifier models (ensemble learning) via stacked generalization of the ESN with leaky integrator neurons approach led to a substantial increase in detection performance of φ of 0.51 and accuracy of 81.2%. This is a substantial improvement of our previous best result of φ = 0.39 on this data with LDA and stacked generalization.

Prediction of microsleeps from EEG: Preliminary results

Brief episodes of momentarily falling asleep - microsleeps - can have fatal consequences, especially in the transportation sector. In this study, the EEG data of eight subjects, while performing a 1-D tracking task, were used to predict imminent microsleeps. A novel algorithm was developed to improve the accuracy of microsleep identification from two independent measures: tracking performance and face-video. The uncertain labels of gold-standard were then pruned out. Additionally, the state of microsleep at 0.25 s ahead was continuously predicted. Log-power spectral features were then extracted from EEG data. The most relevant features were selected by mutual information. Leave-one-subject-out was performed to test the classifier on an independent subject and this procedure was done for all the subjects. Two oversampling methods, synthetic minority oversampling technique (SMOTE) and adaptive sampling (ADASYN), were utilized to improve the training in the presence of imbalanced data. The best average area under the curve of receiver operating characteristic (AUCroc) of 0.90 was achieved using SMOTE oversampling over a 5.25 s window length, with a corresponding geometric mean (GM) of 0.74. ADASYN oversampling achieved the best sensitivity of 0.76 (cf. 0.70 for SMOTE), but with a lower specificity of 0.77 (cf. 0.86 for SMOTE).

Optimal EEG feature selection from average distance between events and non-events

Biosignal classification systems often have to deal with extraneous features, highly imbalanced datasets, and a low SNR. A robust feature selection/reduction method is a crucial step in this process. Sets of artificial data were generated to test a prototype EEG-based microsleep detection system, consisting of a combination of EEG and 2-s bursts of 15-Hz sinusoids of varied signal-to-noise ratios (SNRs) ranging from 16 to 0.03. The balance between events and non-events was varied between evenly balanced and highly imbalanced (e.g., events occurring only 2% of the time). Features were spectral estimates of various EEG bands (e.g., alpha band power) or ratios between them. A total of 34 features for each of the 16 channels yielded a total of 544 features. Five minutes of EEG from a total of eight subjects were used in the generation of the artificial data. Several feature reduction and classifier structures were investigated. Taking only a single feature corresponding to the maximum of average distance between events and non-events (ADEN) on unbalanced data yielded a phi correlation of 0.94 on the mock data with an SNR of 0.3, compared with a phi coefficient of 0.00 for principal component analysis (PCA). ADEN consistently outperformed alternative system configurations, independent of the classifier utilized. While ADEN's high performance may be due to the nature of the artificial dataset, this simulation has demonstrated strong potential compared to other feature selection/reduction methods.

EEG-based event detection using optimized echo state networks with leaky integrator neurons

This study investigates the classification ability of linear and nonlinear classifiers on biological signals using the electroencephalogram (EEG) and examines the impact of architectural changes within the classifier in order to enhance the classification. Consequently, artificial events were used to validate a prototype EEG-based microsleep detection system based around an echo state network (ESN) and a linear discriminant analysis (LDA) classifier. The artificial events comprised infrequent 2-s long bursts of 15 Hz sinusoids superimposed on prerecorded 16-channel EEG data which provided a means of determining and optimizing the accuracy of overall classifier on `gold standard' events. The performance of this system was tested on different signal-to-noise amplitude ratios (SNRs) ranging from 16 down to 0.03. Results from several feature selection/reduction and pattern classification modules indicated that training the classifier using a leaky-integrator neuron ESN structure yielded highest classification accuracy. For datasets with a low SNR of 0.3, training the leaky-neuron ESN using only those features which directly correspond to the underlying event, resulted in a phi correlation of 0.92 compared to 0.37 that employed principal component analysis (PCA). On the same datasets, other classifiers such as LDA and simple ESNs using PCA performed weakly with a correlation of 0.05 and 0 respectively. These results suggest that ESNs with leaky neuron architectures have superior pattern recognition properties. This, in turn, may reflect their superior ability to exploit differences in state dynamics and, hence, provide superior temporal characteristics in learning.

Detecting alpha spindle events in EEG time series using adaptive autoregressive models

Background

Rhythmic oscillatory activity is widely observed during a variety of subject behaviors and is believed to play a central role in information processing and control. A classic example of rhythmic activity is alpha spindles, which consist of short (0.5-2 s) bursts of high frequency alpha activity. Recent research has shown that alpha spindles in the parietal/occipital area are statistically related to fatigue and drowsiness. These spindles constitute sharp changes in the underlying statistical properties of the signal. Our hypothesis is that change point detection models can be used to identify the onset and duration of spindles in EEG. In this work we develop an algorithm that accurately identifies sudden bursts of narrowband oscillatory activity in EEG using techniques derived from change point analysis. Our motivating example is detection of alpha spindles in the parietal/occipital areas of the brain. Our goal is to develop an algorithm that can be applied to any type of rhythmic oscillatory activity of interest for accurate online detection.

Methods

In this work we propose modeling the alpha band EEG time series using discounted autoregressive (DAR) modeling. The DAR model uses a discounting rate to weigh points measured further in the past less heavily than points more recently observed. This model is used together with predictive loss scoring to identify periods of EEG data that are statistically significant.

Results

Our algorithm accurately captures changes in the statistical properties of the alpha frequency band. These statistical changes are highly correlated with alpha spindle occurrences and form a reliable measure for detecting alpha spindles in EEG. We achieve approximately 95% accuracy in detecting alpha spindles, with timing precision to within approximately 150 ms, for two datasets from an experiment of prolonged simulated driving, as well as in simulated EEG. Sensitivity and specificity values are above 0.9, and in many cases are above 0.95, for our analysis.

Conclusion

Modeling the alpha band EEG using discounted AR models provides an efficient method for detecting oscillatory alpha activity in EEG. The method is based on statistical principles and can generally be applied to detect rhythmic activity in any frequency band or brain region.

Distinct neural correlates of time-on-task and transient errors during a visuomotor tracking task after sleep restriction

Sleep loss leads to both time-on-task slowing of responsiveness and increased frequency of transient response errors. The consequences of such errors during real-world visuomotor tasks, such as driving, are serious and life threatening. To investigate the neuronal underpinning of time-on-task and transient errors during a visuomotor tracking task following sleep restriction, we performed fMRI on 20 healthy individuals when well-rested and when sleep-restricted while they performed a 2-D pursuit-tracking task. Sleep restriction to 4-h time-in-bed was associated with significant time-on-task decline in tracking performance and an increased number of transient tracking errors. Sleep restriction was associated with time-on-task decreases in BOLD activity in task-related areas, including the lateral occipital cortex, intraparietal cortex, and primary motor cortex. In contrast, thalamic, anterior cingulate, and medial frontal cortex areas showed overall increases irrespective of time-on-task after sleep-restriction. Furthermore, transient errors after sleep-restriction were associated with distinct transient BOLD activations in areas not involved in tracking task per se, in the right superior parietal cortex, bilateral temporal cortex, and thalamus. These results highlight the distinct cerebral underpinnings of sustained and transient modulations in alertness during increased homeostatic drive to sleep. Ability to detect neuronal changes associated with both sustained and transient changes in performance in a single task allowed us to disentangle neuronal mechanisms underlying two important aspects of sustained task performance following sleep loss.

Measuring neurophysiological signals in aircraft pilots and car drivers for the assessment of mental workload, fatigue and drowsiness

This paper reviews published papers related to neurophysiological measurements (electroencephalography: EEG, electrooculography EOG; heart rate: HR) in pilots/drivers during their driving tasks. The aim is to summarise the main neurophysiological findings related to the measurements of pilot/driver's brain activity during drive performance and how particular aspects of this brain activity could be connected with the important concepts of "mental workload", "mental fatigue" or "situational awareness". Review of the literature suggests that exists a coherent sequence of changes for EEG, EOG and HR variables during the transition from normal drive, high mental workload and eventually mental fatigue and drowsiness. In particular, increased EEG power in theta band and a decrease in alpha band occurred in high mental workload. Successively, increased EEG power in theta as well as delta and alpha bands characterise the transition between mental workload and mental fatigue. Drowsiness is also characterised by increased blink rate and decreased HR values. The detection of such mental states is actually performed "offline" with accuracy around 90% but not online. A discussion on the possible future applications of findings provided by these neurophysiological measurements in order to improve the safety of the vehicles will be also presented.