Feature fusion and WOA-GWO optimization for Alzheimer\u2019s disease detection with sparse EEG channels

Alzheimer's Disease (AD) is a neurodegenerative disorder characterized by energy diffusion and partial disconnection in the brain, with its main feature being an insidious onset and subtle clinical symptoms. Electroencephalogram (EEG) as a primary tool for assessing and aiding in the diagnosis of brain diseases has been widely used in AD detection. Accurate diagnosis is crucial for preventing the transition from early cognitive impairment to AD and providing early treatment for AD patients. This study aims to establish a hybrid model based on the Improved Artificial Fish Swarm Algorithm (IAFS) and Genetic Algorithm (GA)-IAFS-GA, to determine the optimal channel combination for AD detection under multiple EEG signals. Geometric features and complexity features of AD EEG signals were extracted using Second Order Difference Plot (SODP) and entropy analysis across the full frequency band. Subsequently, Pearson correlation was used for feature ranking, selecting the six least correlated features for each channel. The Relief algorithm was then used to fuse these selected features, with one fused feature representing one channel. Based on this, a feature selection optimization algorithm, IAFS-GA, combining the improved artificial fish swarm algorithm and genetic algorithm, was proposed. Finally, the feature combination was input into a Naive Bayes classifier for the identification of AD patients and normal controls. The feature combination was input into a Naive Bayes classifier for the identification of AD patients and normal controls. Using a five-fold cross-validation strategy across the entire frequency band, the classification accuracy reached 93.53%, with a sensitivity of 98.74%, specificity of 98.25%, and an AUC area of 97.82%. This framework can quickly select appropriate brain channels to enhance the efficiency of detecting AD and other neurological diseases. Moreover, it is the first time that an improved artificial fish swarm genetic combination algorithm and SODP features has been used for channel selection in EEG, proving to be an effective method for AD detection. It is based on SODP analysis, entropy analysis, and intelligent algorithms, which can assist clinicians in rapidly diagnosing AD, reducing the misdiagnosis rate of false positives, and expanding our understanding of brain function in patients with neurological diseases.

Brain–computer interfaces (BCI) enable people to communicate with external instruments through brain activity recorded by electroencephalography (EEG). BCI based on motor imagery (MI) can distinguish activation of specific brain regions by decoding EEG signals and then applying them to different situations. Because the activation regions of the brain are specific, using all EEG channels for classification is redundant and may lead to feature confusion and inconvenience for users when applying EEG. Current EEG channel selection methods focus primarily on a single subject and require the data from the subject to generate the chosen channel, which is inconvenient on the application side to determine suitable channels for new subjects. Therefore, this study introduces a novel method for generalized EEG channel selection. Two datasets are used: the BCI competition IV 2a dataset for generating generalized EEG channels and the OpenBMI dataset for validation by numerous subjects. First, the signals from each channel are fed into EEG-Net for classification and ranked by loss to generate optimal EEG channels. Then, the methods of ranking and non-dominated sorting genetic algorithm (NSGA)-II are used to find different combinations of optimal potential differences. Finally, the generalized EEG channels are generated and validated by EEG-Net again. The validation results show that 88.5% of the subjects can be well-classified in one session, including MI-illiteracy, defined by the dataset. The average accuracy is 77.7% and 79.26% in Sessions 1 and 2, using the average channel number around 5, instead of channels from the motor cortex region or all placed EEG channels.

In this paper, experimental neurophysiologic recording and statistical analysis are combined to investigate the nonlinear characteristic and the cognitive function of the brain. Fuzzy approximate entropy and fuzzy sample entropy are applied to characterize the model-based simulated series and electroencephalograph (EEG) series of Alzheimer's disease (AD). The effectiveness and advantages of these two kinds of fuzzy entropy are first verified through the simulated EEG series generated by the alpha rhythm model, including stronger relative consistency and robustness. Furthermore, in order to detect the abnormality of irregularity and chaotic behavior in the AD brain, the complexity features based on these two fuzzy entropies are extracted in the delta, theta, alpha, and beta bands. It is demonstrated that, due to the introduction of fuzzy set theory, the fuzzy entropies could better distinguish EEG signals of AD from that of the normal than the approximate entropy and sample entropy. Moreover, the entropy values of AD are significantly decreased in the alpha band, particularly in the temporal brain region, such as electrode T3 and T4. In addition, fuzzy sample entropy could achieve higher group differences in different brain regions and higher average classification accuracy of 88.1% by support vector machine classifier. The obtained results prove that fuzzy sample entropy may be a powerful tool to characterize the complexity abnormalities of AD, which could be helpful in further understanding of the disease.

Electroencephalogram (EEG) reflect changes in the electrophysiological activity of the brain and can be used in the diagnosis of Alzheimer’s disease (AD). Each EEG channel provides real-time information about the brain, while different EEG channels contain different information about the brain. Using all EEG channel data for AD diagnosis contains redundancy data, leading to increased computation and reduced data analysis accuracy. In this paper, a diagnostic method for AD based on Particle Swarm Optimization (PSO) EEG channel selection and Gated Recurrent Unit (GRU) is proposed. Using EEG energy as the fitness function and PSO to select EEG channels, the redundant information in EEG data is reduced and the accuracy of EEG data analysis is improved. GRU is a special kind of recurrent neural network (RNN) structure. It uses EEG data extracted by the principal component analysis (PCA) feature based on singular value decomposition (SVD) as input to the model. And it has a good advantage in analyzing the time series of EEG. The results show that the classification accuracy of the method in this paper reaches 0.9487, which is higher than the performance of other proposed methods. Compared to the results of using all EEG channel data analysis, the classification accuracy of this method was improved by 0.0757. It shows that the method proposed in this paper can improve the classification accuracy of EEG in AD classification tasks and can be applied to related classification tasks.

Dynamical, causal, and cross-frequency coupling analysis using the electroencephalogram (EEG) has gained significant attention for diagnosing and characterizing neurological disorders. Selecting important EEG channels is crucial for reducing computational complexity in implementing these methods and improving classification accuracy. In neuroscience, measures of (dis) similarity between EEG channels are often used as functional connectivity (FC) features, and important channels are selected via feature selection. Developing a generic measure of (dis) similarity is important for FC analysis and channel selection. In this study, learning of (dis) similarity information within the EEG is achieved using kernel-based nonlinear manifold learning. The focus is on FC changes and, thereby, EEG channel selection. Isomap and Gaussian Process Latent Variable Model (Isomap-GPLVM) are employed for this purpose. The resulting kernel (dis) similarity matrix is used as a novel measure of linear and nonlinear FC between EEG channels. The analysis of EEG from healthy controls (HC) and patients with mild to moderate Alzheimer’s disease (AD) are presented as a case study. Classification results are compared with other commonly used FC measures. Our analysis shows significant differences in FC between bipolar channels of the occipital region and other regions (i.e. parietal, centro-parietal, and fronto-central) between AD and HC groups. Furthermore, our results indicate that FC changes between channels along the fronto-parietal region and the rest of the EEG are important in diagnosing AD. Our results and its relation to functional networks are consistent with those obtained from previous studies using fMRI, resting-state fMRI and EEG.

Electrencephalography (EEG)-based brain-computer interfaces (BCIs) have become popular as EEG is accepted as the simplest and non-invasive neuroimaging modality to record the brain's electrical activity. In the current BCI research context, apart from predicting extremity movements, recent BCI studies have been interested in accurately predicting finger movements of the same hand using different pattern recognition methods over EEG data collected based on motor imagery (MI), through which a mental image of the desired action is generated when a person ideally simulates or imagines carrying out a certain motor task. Although several pattern recognition methods have already been recommended in literature, majority of the studies focusing on classifying five finger movements, were based on study designs that neglected or excluded the idle state of brain (i.e., no mental task state) during which brain does not carry out any MI task. This study design may result in an increasing number of false positives and a significant decrease in the prediction rates and classification performance. Moreover, recent studies have focused on improving prediction performance using complex feature extraction and machine learning algorithms while ignoring comprehensive EEG channels and feature investigation in the prediction of finger movements from EEGs. Therefore, the objectives of this study are threefold: (i) to develop a more viable and practical system to predict the movements of five fingers and the no mental task (NoMT) state from EEG signals (ii) to analyze the effects of the statistical-significance based feature selection method over four different feature domains (nonlinear domain, time-domain, frequency-domain and time-frequency domain) and their combinations, and (iii) to test these feature sets with different and prominent classifiers. In this study, our major goal is not to explore the best machine algorithm performance, but to investigate the best EEG channels and features that can be used in the classification of finger movements. Hence, the comprehensive analysis of the effectiveness of EEG channels and features is performed utilizing a statistically significant feature distribution over 19 EEG channels for each feature set independently. A bulky dataset of electroencephalographic MI for EEG-based BCIs is used in this study. A total of 1102 EEG features supplied from different feature domains have been investigated. Subsequently, these features were tested with eight well-known classifiers, comprising Decision tree, Discriminant analysis, Naive Bayes, Support vector machine, k-nearest neighbor, Ensemble learning, Neural networks, and Kernel approximation. For subject-dependent analysis, the maximum accuracy of 59.17% was obtained using the EEG features that were selected the most (including (i) energy and variance of five frequency bands in frequency-domain feature set, (ii) all feature types in time-domain, time-frequency domain, and nonlinear domain feature sets) and all EEG channels by the Support vector machine algorithm. For subject-independent analysis, the maximum accuracy of 39.30% was obtained using the mostly selected EEG features (which are (i) all feature types excluding the waveform length, average amplitude change value, absolute difference in standard deviation, and slope-change value feature types in time-domain feature set, (ii) the energy and variance values of all frequency bands except gamma frequency band in frequency-domain feature set, (iii) the entropy value of five frequency bands in time-frequency-domain feature set, and (iv) SD 2 and SD 1/SD 2 values where lag = 1 in nonlinear feature set) and EEG channels (which are (i) some definite EEG channels including 2nd, 3rd, 7th, 11th, 13th, 14th, and 15th channels in time-frequency-domain feature set and (ii) all EEG channels in time-domain, frequency-domain, and nonlinear feature sets) by the Support vector machine classifier. Experimental results demonstrate that despite the high-class number, the proposed approach obtained a modest yet considerable advancement in finger movement prediction when the results are compared to the results of similar studies. Additionally, for almost all feature sets, the statistical significance-based feature reduction method improves the prediction performance in the most of classifiers, contributing elaborate EEG channel and feature analysis. Nonetheless, in this study, we used an EEG dataset recorded from only 13 healthy subjects; therefore, a dataset covering more subjects is necessary to reach a more general conclusion.

Basically brain-computer interfaces (BCIs) make use of multichannel electroencephalogram (EEG) devices. Motor imagery tasks are at the core of BCIs and their performance in real-time systems need analysis of huge amounts of signal data coming from multiple channels. The process can be enhanced by selecting a subset of EEG channels to get away with noisy and irrelevant channels. Usage of lesser channels makes the system more convenient to be used in real time applications. This paper deals with the problem of optimal EEG channel selection using enhanced version of differential evolutionary (DE) algorithm. It is straightforward to use the task-specific channels pertaining to the stimulus like \(C_3\), \(C_Z\) and \(C_4\) in case of motor imagery. However, our findings show that the classification performance improves by selecting the optimal channels over the task-specific channels. Also, the performance of the optimal channels is compared with the performance achieved by using all the channels and it is seen that the proposed enhanced variant of DE is beneficial in this regard.

Recently, the electroencephalogram (EEG) signal presents an excellent potential for a new person identification technique. Several studies defined the EEG with unique features, universality, and natural robustness to be used as a new track to prevent spoofing attacks. The EEG signals are a visual recording of the brain's electrical activities, measured by placing electrodes (channels) in various scalp positions. However, traditional EEG-based systems lead to high complexity with many channels, and some channels have critical information for the identification system while others do not. Several studies have proposed a single objective to address the EEG channel for person identification. Unfortunately, these studies only focused on increasing the accuracy rate without balancing the accuracy and the total number of selected EEG channels. The novelty of this paper is to propose a multiobjective binary version of the cuckoo search algorithm (MOBCS-KNN) to find optimal EEG channel selections for person identification. The proposed method (MOBCS-KNN) used a weighted sum technique to implement a multiobjective approach. In addition, a KNN classifier for EEG-based biometric person identification is used. It is worth mentioning that this is the initial investigation of using a multiobjective technique with EEG channel selection problem. A standard EEG motor imagery dataset is used to evaluate the performance of the MOBCS-KNN. The experiments show that the MOBCS-KNN obtained accuracy of 93.86% using only 24 sensors with AR20 autoregressive coefficients. Another critical point is that the MOBCS-KNN finds channels not too close to each other to capture relevant information from all over the head. In conclusion, the MOBCS-KNN algorithm achieves the best results compared with metaheuristic algorithms. Finally, the recommended approach can draw future directions to be applied to different research areas.

Brain-computer interface (BCI) technology has shown the capability of improving the quality of life for people with severe motor disabilities. To improve the portability and practicability of BCI systems, it is crucial to reduce the number of EEG channels as well as to have a good reliability. However, a relatively neglected issue in the EEG channel selection studies is the robustness of selected channels across sessions. This paper investigates whether the selected channels from first session is also useful for subsequent sessions on other days for a stroke patient. For this purpose, a new robust sparse common spatial pattern (RSCSP) algorithm is proposed for optimal EEG channel selection. Thereafter, the robustness of the proposed algorithm as well as 5 existing channel selection algorithms is investigated across 12 sessions data from 11 stroke patients who performed motor imagery based-BCI rehabilitation. The experimental results show that the proposed RSCSP channel selection algorithm significantly outperforms the other channel selection algorithms, when the 8 channels selected from the first session are evaluated on the 11 subsequent sessions. Moreover, there is no significant difference between the classification results of 8 channels selected by the proposed RSCSP algorithm from the first session and the classification results of 8 optimal channels selected from the same session as the test session.

This paper presents a systematic method to select optimal electroencephalography (EEG) channels for three mental tasks-based brain-computer interface (BCI) classification. A blind source separation (BSS) technique based on independent component analysis (ICA) with its back-projecting of the scalp map was used for selecting the optimal EEG channels. The three mental tasks included: mental letter composing, mental arithmetic and mental Rubik's cube rolling. Based on a power spectral density (PSD), the features of the two-channel EEG data were extracted, and then were classified by Bayesian neural network. The results of the ICA decomposition with the back-projected scalp map showed that the prominent channels could be selected for dominant features from original six EEG channels (C3, C4, P3, P4, O1, O2) to four dominant channels (P3, O1, C4, O2) with the best two EEG channels selection at O1C P3&C4 with an accuracy of 71.9% and O1&O2 with an accuracy of 70%.

Dynamical, causal and cross-frequency coupling analysis using the EEG has received significant interest for the analysis and diagnosis of neurological disorders [1]–[3]. Due to the high computational requirements needed for some of these methods, EEG channel selection is crucial [4]. Functional connectivity (FC) between EEG channels is often used for channel selection and connectivity analysis [4, S, 6]. Ideally, in the case of selecting channels for dynamical and causal analysis, FC methods should be able to account for linear and nonlinear spatial and temporal interactions between EEG channels. In neuroscience, FC is quantified using different measures of (dis) similarity to assess the statistical dependence between two signals [5]. However, the interpretation of FC measures can differ significantly from one measure to another[5, 7]. In the early diagnosis of AD, [7] showed correlations among various (dis)similarity measures, and therefore these measures can be grouped. Thus, one from each is sufficient to extract information from the data [7]. Therefore, the development of a generic measure of (dis)similarity is important in FC analysis.

Alzheimer's disease (AD) is a prevalent neurodegenerative disorder affecting millions worldwide. Electroencephalography (EEG), a non-invasive, cost-effective, and safe diagnostic tool, is widely used for detecting neurological conditions. Existing EEG-based classification methods for AD diagnosis have limitations, particularly in adequately considering causal relationships between channels and implementing optimal feature selection, creating a need for highly interpretable feature screening mechanisms. This study presents EENet-RLA, a framework that integrates dynamical system theory with deep learning for AD classification, validated on the BrainLat EEG dataset. The framework operates in two stages, feature extraction and EEG classification, with the deep learning architecture serving primarily as a feature mapping and representation extractor. The core methodological contribution lies in the causal, stability-driven EEG channel selection strategy based on embedding entropy (EE), which quantifies nonlinear directional interactions between EEG channels. This strategy combines bootstrap resampling, multiple random seeds, and minimum connectivity thresholds to identify reproducible, informative channels under limited sample conditions. For classification, spatial and temporal EEG features are extracted using ResNet and LSTM respectively, then fused via a Multi-Head Attention mechanism to capture discriminative patterns. The proposed approach achieves 98.54% segment-level classification accuracy and perfect individual-level performance, demonstrating the discriminative potential of causality-informed feature selection in small-sample settings. While ensuring high accuracy, the method streamlines the analytical process and demonstrates the feasibility of causal-based EEG channel selection in AD characterization, with potential applicability to studying other neurological conditions with similar signal characteristics.

In this study we explore how different levels of emotional intensity (Arousal) and pleasantness (Valence) are reflected in electroencephalographic (EEG) signals. We performed the experiments on EEG data of 32 subjects from the DEAP public dataset, where the subjects were stimulated using 60-s videos to elicitate different levels of Arousal/Valence and then self-reported the rating from 1 to 9 using the self-assessment Manikin (SAM). The EEG data was pre-processed and used as input to a convolutional neural network (CNN). First, the 32 EEG channels were used to compute the maximum accuracy level obtainable for each subject as well as for creating a single model using data from all the subjects. The experiment was repeated using one channel at a time, to see if specific channels contain more information to discriminate between low vs high arousal/valence. The results indicate than using one channel the accuracy is lower compared to using all the 32 channels. An optimization process for EEG channel selection is then designed with the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with the objective to obtain optimal channel combinations with high accuracy recognition. The genetic algorithm evaluates all possible combinations using a chromosome representation for all the 32 channels, and the EEG data from each chromosome in the different populations are tested iteratively solving two unconstrained objectives; to maximize classification accuracy and to reduce the number of required EEG channels for the classification process. Best combinations obtained from a Pareto-front suggests that as few as 8–10 channels can fulfill this condition and provide the basis for a lighter design of EEG systems for emotion recognition. In the best case, the results show accuracies of up to 1.00 for low vs high arousal using eight EEG channels, and 1.00 for low vs high valence using only two EEG channels. These results are encouraging for research and healthcare applications that will require automatic emotion recognition with wearable EEG.

Brain-computer interfaces (BCIs) based on motor imagery (MI) utilizing multi-channel electroencephalogram (EEG) data are commonly used to improve motor function of people with motor disabilities. EEG channel selection can enhance MI classification accuracy by selecting informative channels, accordingly reducing redundant information. The sequential backward floating search (SBFS) approach has been considered as one of the best feature selection methods. In this paper, SBFS is first implemented to select the optimal EEG channels in MI-BCI. Further, to reduce the time complexity of SBFS, the modified SBFS is proposed and applied to left and right hand MI tasks. In the modified SBFS, based on the map of EEG channels at the scalp, the symmetrical channels are selected as channel pairs and acceleration is thus realized by removing or adding multiple channels in each iteration. Extensive experiments were conducted on four public BCI datasets. Experimental results show that the SBFS achieves significantly higher classification accuracy (p < 0.001) than using all channels and conventional MI channels (i.e., C3, C4, and Cz). Moreover, the proposed method outperforms the state-of-the-art selection methods.

Accurate and efficient detection of Alzheimer’s Disease (AD) and Frontotemporal Dementia (FTD) poses significant challenges in neuroscience and healthcare. Our proposed AI framework, Dual-Transformer Cross-Attention Network (DTCA-Net), combines deep learning and directed Phase Transfer Entropy (dPTE) to optimize channel selection crucial for the detection of AD and FTD. Our method identified six critical EEG channels (F7, F8, T3, T4, O1, and O2) through dPTE analysis. We propose the DTCA-Net architecture, which fuses dPTE and differential entropy (DE) features via a multi-head cross-attention layer, projecting dPTE into queries and DE into keys/values over the temporal dimension, so that connectivity (dPTE) shifts are explicitly aligned with spectral complexity (DE), yielding richer spatiotemporal representations. By leveraging a reduced set of EEG channels identified via dPTE, DTCA-Net performance is comparable to previous state-of-the-art models. Additionally, we introduce an adaptive post-processing voting mechanism to enhance subject-level predictions. This approach achieves an F1 score of 84.9% for AD vs. control (CN) detection and 66.5% for FTD vs. CN detection. Overall, compared to traditional full-channel utilization, our model demonstrates the practicality of a reduced-channel solution for clinical applications in AD and FTD detection, enhancing the accessibility and cost-effectiveness of EEG-based diagnostics. The code has been released on GitHub . • Identified key EEG channels for Alzheimer’s detection (AD) using phase transfer entropy. • Introduced a Dual-Transformer Cross-Attention Network using only six EEG channels. • Highlighted lateral frontal, temporal, and occipital electrodes as crucial for AD.