Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine.

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

EEG functional connectivity as a Riemannian mediator: An application to malnutrition and cognition.

Mediation analysis assesses whether an exposure directly produces changes in cognitive behavior or is influenced by intermediate "mediators". Electroencephalographic (EEG) spectral measurements have been previously used as effective mediators representing diverse aspects of brain function. However, it has been necessary to collapse EEG measures onto a single scalar using standard mediation methods. In this article, we overcome this limitation and examine EEG frequency-resolved functional connectivity measures as a mediator using the full EEG cross-spectral tensor (CST). Since CST samples do not exist in Euclidean space but in the Riemannian manifold of positive-definite tensors, we transform the problem, allowing for the use of classic multivariate statistics. Toward this end, we map the data from the original manifold space to the Euclidean tangent space, eliminating redundant information to conform to a "compressed CST." The resulting object is a matrix with rows corresponding to frequencies and columns to cross spectra between channels. We have developed a novel matrix mediation approach that leverages a nuclear norm regularization to determine the matrix-valued regression parameters. Furthermore, we introduced a global test for the overall CST mediation and a test to determine specific channels and frequencies driving the mediation. We validated the method through simulations and applied it to our well-studied 50+-year Barbados Nutrition Study dataset by comparing EEGs collected in school-age children (5-11 years) who were malnourished in the first year of life with those of healthy classmate controls. We hypothesized that the CST mediates the effect of malnutrition on cognitive performance. We can now explicitly pinpoint the frequencies (delta, theta, alpha, and beta bands) and regions (frontal, central, and occipital) in which functional connectivity was altered in previously malnourished children, an improvement to prior studies. Understanding the specific networks impacted by a history of postnatal malnutrition could pave the way for developing more targeted and personalized therapeutic interventions. Our methods offer a versatile framework applicable to mediation studies encompassing matrix and Hermitian 3D tensor mediators alongside scalar exposures and outcomes, facilitating comprehensive analyses across diverse research domains.

Minimum spanning tree analysis of EEG resting-state functional networks in schizophrenia.

Schizophrenia is a serious and complex mental disease, known to be associated with various subtle structural and functional deviations in the brain. Recently, increased attention is given to the analysis of brain-wide, global mechanisms, strongly altering the communication of long-distance brain areas in schizophrenia. Data of 32 patients with schizophrenia and 28 matched healthy control subjects were analyzed. Two minutes long 64-channel EEG recordings were registered during resting, eyes closed condition. Average connectivity strength was estimated with Weighted Phase Lag Index (wPLI) in lower frequencies: delta and theta, and Amplitude Envelope Correlation with leakage correction (AEC-c) in higher frequencies: alpha, beta, lower gamma and higher gamma. To analyze functional network topology Minimum Spanning Tree (MST) algorithms were applied. Results show that patients have weaker functional connectivity in delta and alpha frequency bands. Concerning network differences, the result of lower diameter, higher leaf number, and also higher maximum degree and maximum betweenness centrality in patients suggest a star-like, and more random network topology in patients with schizophrenia. Our findings are in accordance with some previous findings based on resting-state EEG (and fMRI) data, suggesting that MST network structure in schizophrenia is biased towards a less optimal, more centralized organization.

Detecting emotions through EEG signals based on modified convolutional fuzzy neural network.

Emotion is a human sense that can influence an individual's life quality in both positive and negative ways. The ability to distinguish different types of emotion can lead researchers to estimate the current situation of patients or the probability of future disease. Recognizing emotions from images have problems concealing their feeling by modifying their facial expressions. This led researchers to consider Electroencephalography (EEG) signals for more accurate emotion detection. However, the complexity of EEG recordings and data analysis using conventional machine learning algorithms caused inconsistent emotion recognition. Therefore, utilizing hybrid deep learning models and other techniques has become common due to their ability to analyze complicated data and achieve higher performance by integrating diverse features of the models. However, researchers prioritize models with fewer parameters to achieve the highest average accuracy. This study improves the Convolutional Fuzzy Neural Network (CFNN) for emotion recognition using EEG signals to achieve a reliable detection system. Initially, the pre-processing and feature extraction phases are implemented to obtain noiseless and informative data. Then, the CFNN with modified architecture is trained to classify emotions. Several parametric and comparative experiments are performed. The proposed model achieved reliable performance for emotion recognition with an average accuracy of 98.21% and 98.08% for valence (pleasantness) and arousal (intensity), respectively, and outperformed state-of-the-art methods.

FetchEEG: a hybrid approach combining feature extraction and temporal-channel joint attention for EEG-based emotion classification.

Objective. Electroencephalogram (EEG) analysis has always been an important tool in neural engineering, and the recognition and classification of human emotions are one of the important tasks in neural engineering. EEG data, obtained from electrodes placed on the scalp, represent a valuable resource of information for brain activity analysis and emotion recognition. Feature extraction methods have shown promising results, but recent trends have shifted toward end-to-end methods based on deep learning. However, these approaches often overlook channel representations, and their complex structures pose certain challenges to model fitting.Approach. To address these challenges, this paper proposes a hybrid approach named FetchEEG that combines feature extraction and temporal-channel joint attention. Leveraging the advantages of both traditional feature extraction and deep learning, the FetchEEG adopts a multi-head self-attention mechanism to extract representations between different time moments and channels simultaneously. The joint representations are then concatenated and classified using fully-connected layers for emotion recognition. The performance of the FetchEEG is verified by comparison experiments on a self-developed dataset and two public datasets.Main results. In both subject-dependent and subject-independent experiments, the FetchEEG demonstrates better performance and stronger generalization ability than the state-of-the-art methods on all datasets. Moreover, the performance of the FetchEEG is analyzed for different sliding window sizes and overlap rates in the feature extraction module. The sensitivity of emotion recognition is investigated for three- and five-frequency-band scenarios.Significance. FetchEEG is a novel hybrid method based on EEG for emotion classification, which combines EEG feature extraction with Transformer neural networks. It has achieved state-of-the-art performance on both self-developed datasets and multiple public datasets, with significantly higher training efficiency compared to end-to-end methods, demonstrating its effectiveness and feasibility.

Neurovascular Coupling Analysis Based on Multivariate Variational Gaussian Process Convergent Cross-Mapping.

Neurovascular coupling (NVC) provides important insights into the intricate activity of brain functioning and may aid in the early diagnosis of brain diseases. Emerging evidences have shown that NVC could be assessed by the coupling between electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS). However, this endeavor presents significant challenges due to the absence of standardized methodologies and reliable techniques for coupling analysis of these two modalities. In this study, we introduced a novel method, i.e., the collaborative multi-output variational Gaussian process convergent cross-mapping (CMVGP-CCM) approach to advance coupling analysis of EEG and fNIRS. To validate the robustness and reliability of the CMVGP-CCM method, we conducted extensive experiments using chaotic time series models with varying noise levels, sequence lengths, and causal driving strengths. In addition, we employed the CMVGP-CCM method to explore the NVC between EEG and fNIRS signals collected from 26 healthy participants using a working memory (WM) task. Results revealed a significant causal effect of EEG signals, particularly the delta, theta, and alpha frequency bands, on the fNIRS signals during WM. This influence was notably observed in the frontal lobe, and its strength exhibited a decline as cognitive demands increased. This study illuminates the complex connections between brain electrical activity and cerebral blood flow, offering new insights into the underlying NVC mechanisms of WM.

The value of linear and non-linear quantitative EEG analysis in paediatric epilepsy surgery: a machine learning approach.

Epilepsy surgery is effective for patients with medication-resistant seizures, however 20-40% of them are not seizure free after surgery. Aim of this study is to evaluate the role of linear and non-linear EEG features to predict post-surgical outcome. We included 123 paediatric patients who underwent epilepsy surgery at Bambino Gesù Children Hospital (January 2009-April 2020). All patients had long term video-EEG monitoring. We analysed 1-min scalp interictal EEG (wakefulness and sleep) and extracted 13 linear and non-linear EEG features (power spectral density (PSD), Hjorth, approximate entropy, permutation entropy, Lyapunov and Hurst value). We used a logistic regression (LR) as feature selection process. To quantify the correlation between EEG features and surgical outcome we used an artificial neural network (ANN) model with 18 architectures. LR revealed a significant correlation between PSD of alpha band (sleep), Mobility index (sleep) and the Hurst value (sleep and awake) with outcome. The fifty-four ANN models gave a range of accuracy (46-65%) in predicting outcome. Within the fifty-four ANN models, we found a higher accuracy (64.8% ± 7.6%) in seizure outcome prediction, using features selected by LR. The combination of PSD of alpha band, mobility and the Hurst value positively correlate with good surgical outcome.

Detection of Language Lateralization Using Spectral Analysis of EEG.

PURPOSE: Language lateralization relies on expensive equipment and can be difficult to tolerate. We assessed if lateralized brain responses to a language task can be detected with spectral analysis of electroencephalography (EEG). METHODS: Twenty right-handed, neurotypical adults (28 ± 10 years; five males) performed a verb generation task and two control tasks (word listening and repetition). We measured changes in EEG activity elicited by tasks (the event-related spectral perturbation [ERSP]) in the theta, alpha, beta, and gamma frequency bands in two language (superior temporal and inferior frontal [ST and IF]) and one control (occipital [Occ]) region bilaterally. We tested whether language tasks elicited (1) changes in spectral power from baseline (significant ERSP) at any region or (2) asymmetric ERSPs between matched left and right regions. RESULTS: Left IF beta power (-0.37±0.53, t = -3.12, P = 0.006) and gamma power in all regions decreased during verb generation. Asymmetric ERSPs (right > left) occurred between the (1) IF regions in the beta band (right vs. left difference of 0.23±0.37, t(19) = -2.80, P = 0.0114) and (2) ST regions in the alpha band (right vs. left difference of 0.48±0.63, t(19) = -3.36, P = 0.003). No changes from baseline or hemispheric asymmetries were noted in language regions during control tasks. On the individual level, 16 (80%) participants showed decreased left IF beta power from baseline, and 16 showed ST alpha asymmetry. Eighteen participants (90%) showed one of these two findings. CONCLUSIONS: Spectral EEG analysis detects lateralized responses during language tasks in frontal and temporal regions. Spectral EEG analysis could be developed into a readily available language lateralization modality.

A multiscale symbolic approach to decoding delta and ripple oscillation bands as biomarkers for epileptiform discharges.

We use a multiscale symbolic approach to study the complex dynamics of temporal lobe refractory epilepsy employing high-resolution intracranial electroencephalogram (iEEG). We consider the basal and preictal phases and meticulously analyze the dynamics across frequency bands, focusing on high-frequency oscillations up to 240 Hz. Our results reveal significant periodicities and critical time scales within neural dynamics across frequency bands. By bandpass filtering neural signals into delta, theta, alpha, beta, gamma, and ripple high-frequency bands (HFO), each associated with specific neural processes, we examine the distinct nonlinear dynamics. Our method introduces a reliable approach to pinpoint intrinsic time lag scales τ within frequency bands of the basal and preictal signals, which are crucial for the study of refractory epilepsy. Using metrics such as permutation entropy (H), Fisher information (F), and complexity (C), we explore nonlinear patterns within iEEG signals. We reveal the intrinsic τmax that maximize complexity within each frequency band, unveiling the nonlinear subtle patterns of the temporal structures within the basal and preictal signal. Examining the H×F and C×F values allows us to identify differences in the delta band and a band between 200 and 220 Hz (HFO 6) when comparing basal and preictal signals. Differences in Fisher information in the delta and HFO 6 bands before seizures highlight their role in capturing important system dynamics. This offers new perspectives on the intricate relationship between delta oscillations and HFO waves in patients with focal epilepsy, highlighting the importance of these patterns and their potential as biomarkers.

Unsupervised Characterization of Prediction Error Markers in Unisensory and Multisensory Streams Reveal the Spatiotemporal Hierarchy of Cortical Information Processing.

Elicited upon violation of regularity in stimulus presentation, mismatch negativity (MMN) reflects the brain's ability to perform automatic comparisons between consecutive stimuli and provides an electrophysiological index of sensory error detection whereas P300 is associated with cognitive processes such as updating of the working memory. To date, there has been extensive research on the roles of MMN and P300 individually, because of their potential to be used as clinical markers of consciousness and attention, respectively. Here, we intend to explore with an unsupervised and rigorous source estimation approach, the underlying cortical generators of MMN and P300, in the context of prediction error propagation along the hierarchies of brain information processing in healthy human participants. The existing methods of characterizing the two ERPs involve only approximate estimations of their amplitudes and latencies based on specific sensors of interest. Our objective is twofold: first, we introduce a novel data-driven unsupervised approach to compute latencies and amplitude of ERP components accurately on an individual-subject basis and reconfirm earlier findings. Second, we demonstrate that in multisensory environments, MMN generators seem to reflect a significant overlap of "modality-specific" and "modality-independent" information processing while P300 generators mark a shift toward completely "modality-independent" processing. Advancing earlier understanding that multisensory contexts speed up early sensory processing, our study reveals that temporal facilitation extends to even the later components of prediction error processing, using EEG experiments. Such knowledge can be of value to clinical research for characterizing the key developmental stages of lifespan aging, schizophrenia, and depression.

Substance specific EEG patterns in mice undergoing slow anesthesia induction.

The exact mechanisms and the neural circuits involved in anesthesia induced unconsciousness are still not fully understood. To elucidate them valid animal models are necessary. Since the most commonly used species in neuroscience are mice, we established a murine model for commonly used anesthetics/sedatives and evaluated the epidural electroencephalographic (EEG) patterns during slow anesthesia induction and emergence. Forty-four mice underwent surgery in which we inserted a central venous catheter and implanted nine intracranial electrodes above the prefrontal, motor, sensory, and visual cortex. After at least one week of recovery, mice were anesthetized either by inhalational sevoflurane or intravenous propofol, ketamine, or dexmedetomidine. We evaluated the loss and return of righting reflex (LORR/RORR) and recorded the electrocorticogram. For spectral analysis we focused on the prefrontal and visual cortex. In addition to analyzing the power spectral density at specific time points we evaluated the changes in the spectral power distribution longitudinally. The median time to LORR after start anesthesia ranged from 1080 [1st quartile: 960; 3rd quartile: 1080]s under sevoflurane anesthesia to 1541 [1455; 1890]s with ketamine. Around LORR sevoflurane as well as propofol induced a decrease in the theta/alpha band and an increase in the beta/gamma band. Dexmedetomidine infusion resulted in a shift towards lower frequencies with an increase in the delta range. Ketamine induced stronger activity in the higher frequencies. Our results showed substance-specific changes in EEG patterns during slow anesthesia induction. These patterns were partially identical to previous observations in humans, but also included significant differences, especially in the low frequencies. Our study emphasizes strengths and limitations of murine models in neuroscience and provides an important basis for future studies investigating complex neurophysiological mechanisms.

Performance investigation of epilepsy detection from noisy EEG signals using base-2-meta stacking classifier.

Epilepsy is a chronic neurological disease, characterized by spontaneous, unprovoked, recurrent seizures that may lead to long-term disability and premature death. Despite significant efforts made to improve epilepsy detection clinically and pre-clinically, the pervasive presence of noise in EEG signals continues to pose substantial challenges to their effective application. In addition, discriminant features for epilepsy detection have not been investigated yet. The objective of this study is to develop a hybrid model for epilepsy detection from noisy and fragmented EEG signals. We hypothesized that a hybrid model could surpass existing single models in epilepsy detection. Our approach involves manual noise rejection and a novel statistical channel selection technique to detect epilepsy even from noisy EEG signals. Our proposed Base-2-Meta stacking classifier achieved notable accuracy (0.98 ± 0.05), precision (0.98 ± 0.07), recall (0.98 ± 0.05), and F1 score (0.98 ± 0.04) even with noisy 5-s segmented EEG signals. Application of our approach to the specific problem like detection of epilepsy from noisy and fragmented EEG data reveals a performance that is not only superior to others, but also is translationally relevant, highlighting its potential application in a clinic setting, where EEG signals are often noisy or scanty. Our proposed metric DF-A (Discriminant feature-accuracy), for the first time, identified the most discriminant feature with models that give A accuracy or above (A = 95 used in this study). This groundbreaking approach allows for detecting discriminant features and can be used as potential electrographic biomarkers in epilepsy detection research. Moreover, our study introduces innovative insights into the understanding of these features, epilepsy detection, and cross-validation, markedly improving epilepsy detection in ways previously unavailable.

Electroencephalogram-Based Facial Gesture Recognition Using Self-Organizing Map.

Brain-computer interfaces (BCIs) allow information to be transmitted directly from the human brain to a computer, enhancing the ability of human brain activity to interact with the environment. In particular, BCI-based control systems are highly desirable because they can control equipment used by people with disabilities, such as wheelchairs and prosthetic legs. BCIs make use of electroencephalograms (EEGs) to decode the human brain's status. This paper presents an EEG-based facial gesture recognition method based on a self-organizing map (SOM). The proposed facial gesture recognition uses α, ß, and θ power bands of the EEG signals as the features of the gesture. The SOM-Hebb classifier is utilized to classify the feature vectors. We utilized the proposed method to develop an online facial gesture recognition system. The facial gestures were defined by combining facial movements that are easy to detect in EEG signals. The recognition accuracy of the system was examined through experiments. The recognition accuracy of the system ranged from 76.90% to 97.57% depending on the number of gestures recognized. The lowest accuracy (76.90%) occurred when recognizing seven gestures, though this is still quite accurate when compared to other EEG-based recognition systems. The implemented online recognition system was developed using MATLAB, and the system took 5.7 s to complete the recognition flow.

In-Laboratory Polysomnography Worsens Obstructive Sleep Apnea by Changing Body Position Compared to Home Testing.

(1) Background: Home sleep apnea testing, known as polysomnography type 3 (PSG3), underestimates respiratory events in comparison with in-laboratory polysomnography type 1 (PSG1). Without head electrodes for scoring sleep and arousal, in a home environment, patients feel unfettered and move their bodies more naturally. Adopting a natural position may decrease obstructive sleep apnea (OSA) severity in PSG3, independently of missing hypopneas associated with arousals. (2) Methods: Patients with suspected OSA performed PSG1 and PSG3 in a randomized sequence. We performed an additional analysis, called reduced polysomnography, in which we blindly reassessed all PSG1 tests to remove electroencephalographic electrodes, electrooculogram, and surface electromyography data to estimate the impact of not scoring sleep and arousal-based hypopneas on the test results. A difference of 15 or more in the apnea-hypopnea index (AHI) between tests was deemed clinically relevant. We compared the group of patients with and without clinically relevant differences between lab and home tests (3) Results: As expected, by not scoring sleep, there was a decrease in OSA severity in the lab test, similar to the home test results. The group of patients with clinically relevant differences between lab and home tests presented more severe OSA in the lab compared to the other group (mean AHI, 42.5 vs. 20.2 events/h, p = 0.002), and this difference disappeared in the home test. There was no difference between groups in the shift of OSA severity by abolishing sleep scoring in the lab. However, by comparing lab and home tests, there were greater variations in supine AHI and time spent in the supine position in the group with a clinically relevant difference, either with or without scoring sleep, showing an impact of the site of the test on body position during sleep. These variations presented as a marked increase or decrease in supine outcomes according to the site of the test, with no particular trend. (4) Conclusions: In-lab polysomnography may artificially increase OSA severity in a subset of patients by inducing marked changes in body position compared to home tests. The location of the sleep test seems to interfere with the evaluation of patients with more severe OSA.

Classification of Sleep Quality and Aging as a Function of Brain Complexity: A Multiband Non-Linear EEG Analysis.

Understanding and classifying brain states as a function of sleep quality and age has important implications for developing lifestyle-based interventions involving sleep hygiene. Current studies use an algorithm that captures non-linear features of brain complexity to differentiate awake electroencephalography (EEG) states, as a function of age and sleep quality. Fifty-eight participants were assessed using the Pittsburgh Sleep Quality Inventory (PSQI) and awake resting state EEG. Groups were formed based on age and sleep quality (younger adults n = 24, mean age = 24.7 years, SD = 3.43, good sleepers n = 11; older adults n = 34, mean age = 72.87; SD = 4.18, good sleepers n = 9). Ten non-linear features were extracted from multiband EEG analysis to feed several classifiers followed by a leave-one-out cross-validation. Brain state complexity accurately predicted (i) age in good sleepers, with 75% mean accuracy (across all channels) for lower frequencies (alpha, theta, and delta) and 95% accuracy at specific channels (temporal, parietal); and (ii) sleep quality in older groups with moderate accuracy (70 and 72%) across sub-bands with some regions showing greater differences. It also differentiated younger good sleepers from older poor sleepers with 85% mean accuracy across all sub-bands, and 92% at specific channels. Lower accuracy levels (<50%) were achieved in predicting sleep quality in younger adults. The algorithm discriminated older vs. younger groups excellently and could be used to explore intragroup differences in older adults to predict sleep intervention efficiency depending on their brain complexity.

Differentiating Epileptic and Psychogenic Non-Epileptic Seizures Using Machine Learning Analysis of EEG Plot Images.

The treatment of epilepsy, the second most common chronic neurological disorder, is often complicated by the failure of patients to respond to medication. Treatment failure with anti-seizure medications is often due to the presence of non-epileptic seizures. Distinguishing non-epileptic from epileptic seizures requires an expensive and time-consuming analysis of electroencephalograms (EEGs) recorded in an epilepsy monitoring unit. Machine learning algorithms have been used to detect seizures from EEG, typically using EEG waveform analysis. We employed an alternative approach, using a convolutional neural network (CNN) with transfer learning using MobileNetV2 to emulate the real-world visual analysis of EEG images by epileptologists. A total of 5359 EEG waveform plot images from 107 adult subjects across two epilepsy monitoring units in separate medical facilities were divided into epileptic and non-epileptic groups for training and cross-validation of the CNN. The model achieved an accuracy of 86.9% (Area Under the Curve, AUC 0.92) at the site where training data were extracted and an accuracy of 87.3% (AUC 0.94) at the other site whose data were only used for validation. This investigation demonstrates the high accuracy achievable with CNN analysis of EEG plot images and the robustness of this approach across EEG visualization software, laying the groundwork for further subclassification of seizures using similar approaches in a clinical setting.

The Impact of Various Cockpit Display Interfaces on Novice Pilots' Mental Workload and Situational Awareness: A Comparative Study.

Future airspace is expected to become more congested with additional in-service cargo and commercial flights. Pilots will face additional burdens in such an environment, given the increasing number of factors that they must simultaneously consider while completing their work activities. Therefore, care and attention must be paid to the mental workload (MWL) experienced by operating pilots. If left unaddressed, a state of mental overload could affect the pilot's ability to complete his or her work activities in a safe and correct manner. This study examines the impact of two different cockpit display interfaces (CDIs), the Steam Gauge panel and the G1000 Glass panel, on novice pilots' MWL and situational awareness (SA) in a flight simulator-based setting. A combination of objective (EEG and HRV) and subjective (NASA-TLX) assessments is used to assess novice pilots' cognitive states during this study. Our results indicate that the gauge design of the CDI affects novice pilots' SA and MWL, with the G1000 Glass panel being more effective in reducing the MWL and improving SA compared with the Steam Gauge panel. The results of this study have implications for the design of future flight deck interfaces and the training of future pilots.

Identifying the Effect of Cognitive Motivation with the Method Based on Temporal Association Rule Mining Concept.

Being motivated has positive influences on task performance. However, motivation could result from various motives that affect different parts of the brain. Analyzing the motivation effect from all affected areas requires a high number of EEG electrodes, resulting in high cost, inflexibility, and burden to users. In various real-world applications, only the motivation effect is required for performance evaluation regardless of the motive. Analyzing the relationships between the motivation-affected brain areas associated with the task's performance could limit the required electrodes. This study introduced a method to identify the cognitive motivation effect with a reduced number of EEG electrodes. The temporal association rule mining (TARM) concept was used to analyze the relationships between attention and memorization brain areas under the effect of motivation from the cognitive motivation task. For accuracy improvement, the artificial bee colony (ABC) algorithm was applied with the central limit theorem (CLT) concept to optimize the TARM parameters. From the results, our method can identify the motivation effect with only FCz and P3 electrodes, with 74.5% classification accuracy on average with individual tests.

Calibrating Deep Learning Classifiers for Patient-Independent Electroencephalogram Seizure Forecasting.

The recent scientific literature abounds in proposals of seizure forecasting methods that exploit machine learning to automatically analyze electroencephalogram (EEG) signals. Deep learning algorithms seem to achieve a particularly remarkable performance, suggesting that the implementation of clinical devices for seizure prediction might be within reach. However, most of the research evaluated the robustness of automatic forecasting methods through randomized cross-validation techniques, while clinical applications require much more stringent validation based on patient-independent testing. In this study, we show that automatic seizure forecasting can be performed, to some extent, even on independent patients who have never been seen during the training phase, thanks to the implementation of a simple calibration pipeline that can fine-tune deep learning models, even on a single epileptic event recorded from a new patient. We evaluate our calibration procedure using two datasets containing EEG signals recorded from a large cohort of epileptic subjects, demonstrating that the forecast accuracy of deep learning methods can increase on average by more than 20%, and that performance improves systematically in all independent patients. We further show that our calibration procedure works best for deep learning models, but can also be successfully applied to machine learning algorithms based on engineered signal features. Although our method still requires at least one epileptic event per patient to calibrate the forecasting model, we conclude that focusing on realistic validation methods allows to more reliably compare different machine learning approaches for seizure prediction, enabling the implementation of robust and effective forecasting systems that can be used in daily healthcare practice.