The scientific literature highlights the significance of human motor variability in understanding motor control, learning, and neurological disorders. Visuomotor tasks in laboratory settings offer a controlled platform for studying motor variability, but the specialized equipment used for these tasks limits their accessibility and generalizability. Thus, this study aimed to develop an accessible, standardized mouse-based task for home-based assessment of motor variability characteristics. Two protocols were conducted: (1) comparing the mouse-based task to a well-established force-sensor lab task (N = 10), and (2) assessing the mouse-based task across various computer setups (N = 31). Results demonstrated high reliability and accuracy of both tasks for the following measures of motor variability: fuzzy entropy and detrended fluctuation analysis (DFA). The mouse-based task exhibited slightly superior absolute reliability, suggesting potential sensitivity advantages for detecting longitudinal changes in motor control. While sampling frequency influenced nonlinear outputs, it did not significantly affect reliability, leading to the choice of 20Hz for optimal parameter estimation. Correlation analyses revealed that although participants showed different performance during the mouse- and force-sensor-based tasks, their long-term movement adjustment strategies (assessed using the DFA) were similar. In addition, the robustness analysis showed that computer hardware can influence observed variability, with screen size being a key factor. Larger screens may increase error sensitivity and affect variability structure. Overall, the findings highlight the potential of the mouse-based task for home-based motor variability assessment, emphasizing its reliability, accuracy, and adaptability across various computer setups.

AI based intuitionistic dense fuzzy entropy decision- making system and its application in firefighting robot selection

Non-linear neural dynamics reflect the inherent complexity of brain activity and are increasingly recognized as important indicators of neural adaptability and integrity. Bipolar disorder (BD) is associated with atypical brain activity, as evidenced by prior research demonstrating altered electroencephalography (EEG) spectral entropy and gamma-band auditory steady-state response (aSSR) deficits, suggesting impaired neural adaptability. In this study, EEG data from 230 participants (BD=90, control=140) were analyzed to investigate differences in signal complexity using fuzzy entropy (FuzzEn) levels during 40Hz auditory entrainment exposure and to evaluate the diagnostic utility of FuzzEn features using an extreme gradient boosting (XGBoost) machine learning classifier. Individuals with BD demonstrated significantly higher FuzzEn at baseline (Median=0.234 vs. 0.142, p<.001) and during stimulus exposure (Median=0.267 vs. 0.196, p<.001), along with reduced entropy modulation (Median=0.023 vs. 0.036, p<.001) compared to controls. The XGBoost classifier achieved a modest diagnostic accuracy of 67%, highlighting the contribution of FuzzEn features in capturing neural dynamics relevant to BD. These findings suggest that impaired neural adaptation to sensory input in BD may be linked to heightened disorder in brain activity, informing the development of diagnostics and therapeutic interventions for BD.

Repetitive Transcranial Magnetic Stimulation (rTMS) shows promise for treating Autism Spectrum Disorder (ASD), but its impact on the temporal dynamics of large-scale brain networks remains unclear. This study investigated the modulatory effects of rTMS on static and dynamic brain functional networks in children with ASD. Thirty-two children were randomized into an active rTMS group (1Hz over the dorsolateral prefrontal cortex) and a sham control group. Resting-state EEG and behavioral assessments were conducted before and after a 9-week intervention. We employed a multi-dimensional analysis approach, combining microstate temporal parameters, static functional connectivity based on the weighted Phase Lag Index (wPLI), and dynamic complexity measured by Fuzzy Entropy. Results indicated that intrinsic features of Microstate B were significantly correlated with social relating deficits. Although rTMS did not induce significant interaction effects in standard microstate temporal parameters, it significantly enhanced static functional connectivity strength and increased the dynamic complexity of brain networks across all microstates. These findings suggest that rTMS exerts its therapeutic effects by strengthening network integration and restoring neural flexibility rather than simply altering the duration of brain states. The study underscores the value of network-based EEG metrics in elucidating the neuroplastic changes induced by neuromodulation in ASD.

Alzheimer’s Disease (AD) is a neurodegenerative disorder with insidious onset, making early diagnosis challenging. Electroencephalogram (EEG) is a promising noninvasive tool for AD diagnosis, but high-density EEG configurations cause computational burdens and hinder clinical translation. Thus, developing an efficient sparse EEG channel selection method with high classification accuracy is urgent for AD auxiliary diagnosis. This study proposes a multi-strategy enhanced Whale Optimization Algorithm-Grey Wolf Optimizer (WOA-GWO) hybrid model for EEG channel selection, combined with a nonlinear dynamic feature fusion framework. We extracted geometric features from second-order difference plot (SODP) and complexity features (sample entropy, fuzzy entropy) of EEG signals, then adopted the ReliefF algorithm for feature fusion and key feature selection. The WOA-GWO model was improved via chaotic initialization, nonlinear convergence factors, spiral-hierarchical position update, and random perturbation to avoid local optima. Experimental results show that the proposed framework achieves a classification accuracy of 96.97% for AD detection, with significantly reduced EEG channel dimensions (four optimal channels identified: T5, FP1, T4, F4). The WOA-GWO model outperforms the original WOA and GWO in convergence speed and optimization accuracy, and the fused features exhibit strong discriminability for AD-related EEG abnormalities. This work provides a reliable computational framework for developing lightweight, portable AD diagnostic systems, and the identified optimal EEG channels offer neurophysiological evidence for AD electrophysiological biomarkers.

Depression is one of the most prevalent mental disorders globally, severely affecting individuals' emotional, cognitive, and physical functions while imposing profound socioeconomic impacts. Traditional diagnostic approaches primarily rely on clinical judgment and self-assessment scales; however, these methods carry inherent risks of misdiagnosis and missed diagnosis, necessitating more precise and efficient diagnostic tools. This study employs a two-channel frontal EEG system for depression detection, aiming to simplify data acquisition processes and reduce costs while ensuring high classification accuracy. Electroencephalography (EEG), as a non-invasive biosignal monitoring technique, enables real-time recording of brain electrical activity. By extracting multiple features including relative power, fuzzy entropy, and mutual information, combined with multi-scale analysis techniques, the detection accuracy for depression was further enhanced. The study compared three traditional machine learning models with three deep learning models, among which the Gated Recurrent Unit (GRU) model demonstrated superior performance, achieving a classification accuracy of 91.02% and exhibiting strong robustness. The aforementioned approach provides preliminary technical support for the application of EEG signals in depression detection, and represents a proof-of-concept for multi-scale feature-enhanced automated depression screening. Further validation in larger, clinically representative, and externally verified cohorts is necessary before practical deployment can be considered.

Two-dimensional fuzzy dispersion entropy to nonlinear image analysis

From resource to innovation: A decision framework for sustainable boron research infrastructure

The use of multi-dimensional information technology to process auditory synesthesia in musical emotions has gradually become a hot topic of interest among many scholars. This paper proposes a novel method for decoding electroencephalography (EEG) signals by combining spectral summation analysis (SSA) with entropy measures. EEG time series are decomposed and reconstructed using SSA to obtain SSA components of various orders. Four entropy measure feature data—approximate entropy, sample entropy, fuzzy entropy, and multiscale entropy—are extracted from the SSA components. These entropy measure feature data are then used to construct a feature vector describing the relevant EEG data decoding recognition, and combined with a pattern classifier to achieve EEG decoding. Additionally, a TCM-CSP emotional EEG analysis method based on cognitive topological constraints is proposed. Considering the temporal changes in emotions, an emotional brain region analysis method is proposed, and combined with the spatial characteristics of EEG signals, a CSP computation method that preserves topological constraints is proposed. Finally, experiments show that in emotional tendency analysis, the distortion group and the reverb group exhibit negative correlations. The distortion group tends toward restlessness and anxiety, while the reverb group tends toward calmness and relaxation, indicating that the tonal quality of sound after different effects processing significantly influences emotional perception.

Abstract Early detection and proper prognosis are important in the treatment of breast cancer as it is one of the major causes of death in women across the world. Traditional diagnostic and predictive models typically do not cope with uncertainty and imprecision of clinical data, which leads to the decrease in reliability in practice. This paper seeks to explore the use of fuzzy-based data mining methods to improve both the prognosis and diagnosis of breast cancer. The research design is quantitative predictive research design based on 250 simulated patient records comprised of tumor characteristics, uniformity of cells, mitotic rate and other diagnostic markers. Preprocessing of key clinical variables is done and encoded into linguistic terms then put through a feature selection process using correlation analysis and fuzzy entropy measures to maximize model efficiency. Development of a fuzzy-based classifier is achieved by fuzzification, rule generation, inference, and defuzzification, which allows fine-tuning of disease presence (Benign/Malignant) and prognosis risk (Low, Moderate, High). The evaluation of performance is performed on the basis of 10-fold cross-validation, and the measures of accuracy, sensitivity, specificity, precision, and F1-score are compared to the traditional machine learning models: Artificial Neural Networks, Support Vector Machines, and Decision Trees. These findings indicate that the fuzzy-based classifier performs better than traditional models and the accuracy of the fuzzy-based classifier is 94 and 92 in diagnosis and prognosis respectively with high sensitivity, specificity, precision, and F1-scores in both tasks. The confusion analysis also proves the strength of the model as there are few misclassifications and solid predictions in all the risk categories. The results suggest that fuzzy-based methods are useful to deal with overlapping and imprecise clinical features, to provide interpretable and actionable clinical decision-making. Lastly, the study reaffirms the fuzzy-based data mining as a robust, flexible and clinically feasible model of breast cancer diagnosis and prognosis, which allows one to diagnose a patient early, assess the risks properly and treat him/her individually. This approach can be extended to future studies using large real patient datasets and incorporating ensemble learning or real-time monitoring in order to realize more predictive capability and clinical utility.

Spontaneous Leg Movements Measured by Wearable Sensors in Infancy Differentiate Later Risk for Cerebral Palsy.

This study investigates the challenging task of predicting the strength of subgrade soils, which serve as the foundation of superstructure systems. Due to the inherent complexity of soil behavior, traditional empirical methods often fall short in providing consistent and reliable estimations. To address this limitation, a fuzzy entropy-based TOPSIS multi-criteria decision-making (MCDM) approach is proposed. Methodologically, the study introduces a novel fuzzy entropy function that extends existing fuzzy entropy formulations and is compared against conventional fuzzy entropy measures. Using the newly proposed Pm− fuzzy entropy (m = 0.5), a soil stabilization quality ranking was obtained and validated against classical fuzzy entropy-based TOPSIS results. It is important to emphasize that the primary objective of the proposed framework is not to provide direct numerical estimates of CBR values, but rather to support the decision-making process by ranking soil options based on multiple criteria under conditions of uncertainty. The robustness of the rankings was further examined using California Bearing Ratio (CBR) data and comprehensive sensitivity analyses to consider uncertainties from expert judgments and laboratory measurements. The proposed approach offers a solution for multi-criteria decision-making processes in uncertain environments, ensuring high rating consistency and adaptability.

BackgroundFrequent, sustained stress is linked to poor health and requires monitoring for early intervention. Electrocardiograms (ECG) are promising biomarkers because they can be recorded noninvasively and continuously using wearable devices. However, tracking stress with ECG is challenging because daily activities elicit responses similar to mental stress (MS), and various mental stimuli that individuals encounter complicate the use of machine learning (ML) models trained on a limited set of stressors.ObjectiveWe (1) evaluated the ability of ML models to distinguish MS episodes from a composite “no-stress” background, including rest and low- to moderate-intensity activities; (2) assessed their generalizability to new stressors and participants; and (3) tested robustness to lower sampling rates and fewer features, to explore their suitability for lightweight wearables.MethodsWe used a comprehensive ECG dataset sampled at 1000 hertz from 127 participants who underwent various mental stressors and engaged in diverse physical activities. A 30-second window was used to extract 55 features from time, frequency, nonlinear, and morphological domains. We trained a logistic regression (LR) model and an extreme gradient boosting (XGBoost) model, splitting the data into 60/20/20 for training, validation, and testing. Shapley additive explanation values were computed to explain model predictions. Additional analyses included leave-one-stressor-out; downsampling to 500, 250, and 125 hertz; a time-window sensitivity analysis; and reducing the number of features to as few as 5.ResultsXGBoost achieved an area under the receiver operating characteristic curve (AUROC) of 0.741 (95% CI 0.701‐0.783) and an area under the precision-recall curve (AUPRC) of 0.706 (95% CI 0.658‐0.753), compared with 0.724 (95% CI 0.678‐0.772) and 0.691 (95% CI 0.639‐0.742) for LR. The mean performance difference between XGBoost and LR was 0.017 for AUROC (95% CI 0.001‐0.032) and 0.015 for AUPRC (95% CI −0.001 to 0.037; clustered bootstrap analysis using 2000 participant-level resamples), suggesting that LR performs comparably to the nonlinear XGBoost model. Both models were robust to downsampling and feature reduction (10 features retained >93% of performance). Extending the analysis window to 60 seconds improved model performance across all sampling rates, highlighting a trade-off between rapid detection and overall performance. When evaluating discrimination from physical activity, models achieved acceptable specificity for light physical activity (XGBoost: 0.787; LR: 0.794) but poor specificity for moderate physical activity (XGBoost: 0.418; LR: 0.444). Both models generalized to most unseen stressors, although performance varied across stressors, with limited transfer to the social-evaluative stressor. Feature importance analysis revealed fuzzy entropy and frequency-based features as key predictors.ConclusionsML models can detect MS with high sensitivity and remain robust to lower sampling rates and fewer features. Generalization to novel stressors was stressor-dependent. Importantly, our results highlight challenges in distinguishing stress-related cardiac responses from those caused by physical exertion, revealing critical limitations of single-sensor ECG approaches for MS detection.

ABSTRACT Air pollution is responsible for various health issues, including respiratory and cardiovascular diseases, among individuals. However, previous studies have not successfully identified the sources of air pollution that contribute to the acceleration of climate change. To address this gap, a novel approach known as the Adaptive Exponential Sigmoid Fuzzy Tsallis Entropy Interference System (AES-FTEIS) is proposed for identifying air pollution sources. Datasets from remote sensing and ground-level air pollution measurements are collected, temporally aligned using the Prior Distribution Regularized Kalman Filter (PDRKF), and imputed using Cross-Entropy Minimization Spline Interpolation (CEMSI). Additionally, aerosol particles such as PM2.5 and PM10 are extracted from the dataset and incorporated into the analysis. Subsequently, the data are organized by location and time using Transfer Entropy Spectral Clustering (TESC), and their correlation are analysed using Spearman Rank Correlation (SRC). An exploratory data analysis is conducted on the time-based grouped results through Box plots, leading to feature extraction. Finally, the AES-FTEIS is utilized to identify the pollution sources based on the levels of aerosol particle concentrations. The experimental results show that the proposed WOLSTM-ASLRCNN classifier achieves 97.56% of accuracy and 98.5% of precision, outperforming existing models such as CNN (95.3%), LSTM (93.84%), GRU (92.57%), and RNN (89.66%). The proposed TESC clustering method obtained a silhouette score of 0.9721, higher than SC (0.9687), AC (0.9428), HC (0.9271) and KMC (0.9087). Moreover, the AES-FTEIS source identification approach reduced the rule generation time to 1483 ms, demonstrating the effectiveness of the proposed framework.

Abstract This paper develops a hybrid fuzzy multi-criteria decision-making (MCDM) framework to support the sustainable selection of hydrogen fuel cell (FC) technologies for combined heat and power (CHP) applications. The approach addresses limitations in prior studies, including reliance on subjective weighting, inadequate treatment of uncertainty, and use of singular ranking techniques. Two objective weighting methods, Fuzzy Distance Correlation–based Criteria Importance Through Inter-Criteria Correlation (Fuzzy-D-CRITIC) and Fuzzy Entropy, are applied to determine criteria importance based on data characteristics, and their respective weights are integrated through a game-theoretic aggregation mechanism to obtain balanced weighting coefficients. Four fuzzy MCDM methods, Combined Compromise Solution (Fuzzy-CoCoSo), Technique for Order Preference by Similarity to Ideal Solution (Fuzzy-TOPSIS), Fuzzy Weighted aggregated sum product assessment (Fuzzy-WASPAS), Fuzzy Complex proportional assessment (Fuzzy-COPRAS) are then employed to evaluate five FC alternatives across seventeen techno-economic and environmental criteria, and their performance scores are aggregated using a fuzzy game-theoretic weighted average to generate a consolidated ranking. The results consistently identify solid oxide fuel cells (SOFCs) as the most suitable technology for CHP applications, owing to their high electrical efficiency, effective heat utilization, and fuel flexibility, while alkaline fuel cells (AFCs) are found to be least suitable. Sensitivity and comparative analyses confirm the robustness and stability of the final rankings against perturbations in criteria weights and aggregation coefficients, as well as their alignment with consensus-based methods. The study provides a structured, data-driven decision-support tool for stakeholders in energy planning, investment, and policy formulation, advancing the selection of sustainable FC technologies for decarbonized energy systems.

UAMOD: Unsupervised feature selection algorithm using adaptive multi-neighborhood fuzzy entropy for outlier detection

Neural responses to acute hypoxia and hyperoxia.

Adaptive composite multi-scale weighted fuzzy entropy: An effective nonlinear dynamic tool for fault feature extraction of rolling bearings

Abstract The core of bearing fault diagnosis lies in the accurate feature extraction from non-stationary vibration signals; however, unavoidable multi-source engineering noise often leads to severe feature distortion. To address this issue, this paper proposes a two-stage robust feature extra-ction framework based on Adaptive Time-Varying Filtering Fast Ensemble Empirical Mode Decomposition (ATVF-FEEMD) and Hierarchical Time-Shift Composite Multi-scale Fuzzy Entropy (HTSCMFE). First, by introducing an adaptive filter and an "information-energy" adaptive dual-criteria screening strategy (Log-MIC and Log-Gini index), ATVF-FEEMD effectively eliminates noise components, achieving high-fidelity signal reconstruction under intense background noise. Second, through the hierarchical decomposition, multi-offset time-shifting, and composite multi-scale coarse-graining of the reconstructed signal via HTSCMFE, the derived feature vectors exhibit outstanding discriminative capability. Experiments on simulated and real-world datasets demonstrate that, owing to the excepti-onally high quality of the extracted features, the proposed framework maintains excellent recognition accuracy and generalization ability across varying Signal-to-Noise Ratios (SNRs), even when utilizing only fundamental classifiers such as Artificial Neural Networks (ANN) and Transformers. Furthermore, this study draws a crucial conclusion: high-quality front-end features can significantly reduce the algorithmic dependency of diagnostic models on complex graph topologies (e.g., GCN/GAT), thereby providing a highly robust new paradigm for bearing fault diagnosis under severe noise conditions.

Accurate forecasting of copper futures prices is crucial for risk management and investment decisions. However, existing approaches primarily rely on historical prices and incorporate behavioral signals without a unified modeling framework. To address this limitation, we propose MBTI-Net (Multi-source Behavior-Triggered Interaction Network), a behavior-aware forecasting framework for heterogeneous copper market data. We first construct a compact behavioral factor from Baidu search indices via a multi-view projection strategy that preserves structural and predictive information. We then develop a complexity-aware reconstruction mechanism that aggregates intrinsic mode functions into multi-frequency components based on fuzzy entropy and energy. To accommodate distributional and volatility differences between behavioral and market variables, we introduce VB-ReVIN (Volatility- and Behavior-aware Reversible Instance Normalization). Building upon these representations, MBTI-Net models dynamic multi-source interactions triggered by behavioral intensity and market conditions, enabling adaptive cross-source information fusion. Experiments on LME and SHFE copper futures datasets demonstrate consistent improvements over state-of-the-art benchmarks, highlighting the importance of explicitly modeling behavior-driven dependencies in financial forecasting.