Subject-specific mental workload classification using EEG and stochastic configuration network (SCN)

Evidence for modulation of EEG microstates by mental workload levels and task types

Electroencephalography (EEG) microstate analysis has become a popular tool for studying the spatial and temporal dynamics of large-scale electrophysiological activities in the brain in recent years. Four canonical topographies of the electric field (classes A, B, C, and D) have been widely identified, and changes in microstate parameters are associated with several psychiatric disorders and cognitive functions. Recent studies have reported the modulation of EEG microstate by mental workload (MWL). However, the common practice of evaluating MWL is in a specific task. Whether the modulation of microstate by MWL is consistent across different types of tasks is still not clear. Here, we studied the topographies and dynamics of microstate in two independent MWL tasks: NBack and the multi-attribute task battery (MATB) and showed that the modulation of MWL on microstate topographies and parameters depended on tasks. We found that the parameters of microstates A and C, and the topographies of microstates A, B, and D were significantly different between the two tasks. Meanwhile, all four microstate topographies and parameters of microstates A and C were different during the NBack task, but no significant difference was found during the MATB task. Furthermore, we employed a support vector machine recursive feature elimination procedure to investigate whether microstate parameters were suitable for MWL classification. An averaged classification accuracy of 87% for within-task and 78% for cross-task MWL discrimination was achieved with at least 10 features. Collectively, our findings suggest that topographies and parameters of microstates can provide valuable information about neural activity patterns with a dynamic temporal structure at different levels of MWL, but the modulation of MWL depends on tasks and their corresponding functional systems. Moreover, as a potential indicator, microstate parameters could be used to distinguish MWL.

Effects of mental workload on manufacturing systems employees: A mediation causal model

BACKGROUND

Although some research has been done in the Mexican manufacturing industry regarding mental workload, none has explored its association with physical fatigue, body weight gain, and human error simultaneously.

OBJECTIVE

This research examines the association between mental workload and physical fatigue, body weight gain, and human error in employees from the Mexican manufacturing systems through a mediation analysis approach.

METHODS

A survey named Mental Workload Questionnaire was developed by merging the NASA-TLX with a questionnaire containing the mental workload variables mentioned above. The Mental Workload Questionnaire was applied to 167 participants in 63 manufacturing companies. In addition, the mental workload was used as an independent variable, while physical fatigue and body weight gain were mediator variables, and human error was a dependent variable. Six hypotheses were used to measure the relationships among variables and tested using the ordinary least squares regression algorithm.

RESULTS

Findings indicated that mental workload significantly correlates with physical fatigue and human error. Also, the mental workload had a significant total association with human error. The highest direct association with body weight gain was provided by physical fatigue, and body weight gain had an insignificant direct association with human error. Finally, all indirect associations were insignificant.

CONCLUSION

Mental workload directly affects human error, which physical fatigue does not; however, it does affect body weight gain. Managers should reduce their employees' mental workload and physical fatigue to avoid further problems associated with their health.

Decoding working memory-related information from repeated psychophysiological EEG experiments using convolutional and contrastive neural networks

OBJECTIVE

Extracting reliable information from EEG signals is difficult because the low signal-to-noise ratio and significant intersubject variability seriously hinder statistical analyses. However, recent advances in explainable machine learning open a new strategy to address this problem.

APPROACH

The current study evaluates this approach using results from the classification and decoding of electrical brain activity associated with information retention. We designed four neural network models differing in architecture, training strategies, and input representation to classify single experimental trials of a working memory task.

MAIN RESULTS

Our best models achieved an accuracy of 65.29$±0.76 and Matthews correlation coefficient of 0.288±0.018, outperforming the reference model trained on the same data. The highest correlation between classification score and behavioral performance was 0.36 (p=0.0007). Using analysis of input perturbation, we estimated the importance of EEG channels and frequency bands in the task at hand. The set of essential features identified for each network varies. We identified a subset of features common to all models that identified brain regions and frequency bands consistent with current neurophysiological knowledge of the processes critical to attention and working memory. Finally, we proposed sanity checks to examine further the robustness of each model's set of features.

SIGNIFICANCE

Our results indicate that explainable deep learning is a powerful tool for decoding information from EEG signals. It is crucial to train and analyze a range of models to identify stable and reliable features. Our results highlight the need for explainable modeling as the model with the highest accuracy appeared to use residual artifactual activity.

Fault Diagnosis of Power Transformer Based on Time-Shift Multiscale Bubble Entropy and Stochastic Configuration Network

In order to accurately diagnose the fault type of power transformer, this paper proposes a transformer fault diagnosis method based on the combination of time-shift multiscale bubble entropy (TSMBE) and stochastic configuration network (SCN). Firstly, bubble entropy is introduced to overcome the shortcomings of traditional entropy models that rely too heavily on hyperparameters. Secondly, on the basis of bubble entropy, a tool for measuring signal complexity, TSMBE, is proposed. Then, the TSMBE of the transformer vibration signal is extracted as a fault feature. Finally, the fault feature is inputted into the stochastic configuration network model to achieve an accurate identification of different transformer state signals. The proposed method was applied to real power transformer fault cases, and the research results showed that TSMBE-SCN achieved 99.01%, 99.1%, 99.11%, 99.11%, 99.14% and 99.02% of the diagnostic rates under different folding numbers, respectively, compared with conventional diagnostic models MBE-SCN, TSMSE-SCN, MSE-SCN, TSMDE-SCN and MDE-SCN. This comparison shows that TSMBE-SCN has a strong competitive advantage, which verifies that the proposed method has a good diagnostic effect. This study provides a new method for power transformer fault diagnosis, which has good reference value.

Mental Workload Classification Method Based on EEG Cross-Session Subspace Alignment

Electroencephalogram (EEG) signals are sensitive to the level of Mental Workload (MW). However, the random non-stationarity of EEG signals will lead to low accuracy and a poor generalization ability for cross-session MW classification. To solve this problem of the different marginal distribution of EEG signals in different time periods, an MW classification method based on EEG Cross-Session Subspace Alignment (CSSA) is presented to identify the level of MW induced in visual manipulation tasks. The Independent Component Analysis (ICA) method is used to obtain the Independent Components (ICs) of labeled and unlabeled EEG signals. The energy features of ICs are extracted as source domains and target domains, respectively. The marginal distributions of source subspace base vectors are aligned with the target subspace base vectors based on the linear mapping. The Kullback–Leibler (KL) divergences between the two domains are calculated to select approximately similar transformed base vectors of source subspace. The energy features in all selected vectors are trained to build a new classifier using the Support Vector Machine (SVM). Then it can realize MW classification using the cross-session EEG signals, and has good classification accuracy.

Subject-Specific Cognitive Workload Classification Using EEG-Based Functional Connectivity and Deep Learning

Cognitive workload is a crucial factor in tasks involving dynamic decision-making and other real-time and high-risk situations. Neuroimaging techniques have long been used for estimating cognitive workload. Given the portability, cost-effectiveness and high time-resolution of EEG as compared to fMRI and other neuroimaging modalities, an efficient method of estimating an individual’s workload using EEG is of paramount importance. Multiple cognitive, psychiatric and behavioral phenotypes have already been known to be linked with “functional connectivity”, i.e., correlations between different brain regions. In this work, we explored the possibility of using different model-free functional connectivity metrics along with deep learning in order to efficiently classify the cognitive workload of the participants. To this end, 64-channel EEG data of 19 participants were collected while they were doing the traditional n-back task. These data (after pre-processing) were used to extract the functional connectivity features, namely Phase Transfer Entropy (PTE), Mutual Information (MI) and Phase Locking Value (PLV). These three were chosen to do a comprehensive comparison of directed and non-directed model-free functional connectivity metrics (allows faster computations). Using these features, three deep learning classifiers, namely CNN, LSTM and Conv-LSTM were used for classifying the cognitive workload as low (1-back), medium (2-back) or high (3-back). With the high inter-subject variability in EEG and cognitive workload and recent research highlighting that EEG-based functional connectivity metrics are subject-specific, subject-specific classifiers were used. Results show the state-of-the-art multi-class classification accuracy with the combination of MI with CNN at 80.87%, followed by the combination of PLV with CNN (at 75.88%) and MI with LSTM (at 71.87%). The highest subject specific performance was achieved by the combinations of PLV with Conv-LSTM, and PLV with CNN with an accuracy of 97.92%, followed by the combination of MI with CNN (at 95.83%) and MI with Conv-LSTM (at 93.75%). The results highlight the efficacy of the combination of EEG-based model-free functional connectivity metrics and deep learning in order to classify cognitive workload. The work can further be extended to explore the possibility of classifying cognitive workload in real-time, dynamic and complex real-world scenarios.