This study presents results from a spatial modeling and analysis process of six different air pollutants measured over a ten-year period at up to forty-three monitoring stations located in the three provinces of the Basque Autonomous Community (BAC) (Spain). The main objective was to generate detailed maps showing the evolution of these pollutants that cover the entire area using geostatistical techniques. These maps are intended to serve as a basis for both short-term and medium-term environmental studies, while also examining how pollutant levels have changed before and after the COVID-19 pandemic era. Additionally, the paper explores the factors that may explain the differences observed during these two periods. To further analyze the spatial patterns, the study employs the Fuzzy C-Means clustering algorithm to partition the region into four distinct zones based on the concentrations of key pollutants: PM10, NO2, and O3. These pollutants were selected due to their high sampling density, spatial coverage, and complementary sources (traffic emissions, combustion processes, and ozone photochemistry), making them representative indicators of the region’s atmospheric state. The findings reveal significant changes in air quality during the COVID-19 pandemic, particularly in NO2, Benzene, and CO levels, which sharply declined due to reduced vehicular traffic. However, the behavior of PM10 and O3 was more complex, influenced by diverse sources and atmospheric chemistry. The reduction in NO2 emissions during lockdowns led to a counterintuitive increase in O3 concentrations in some areas, highlighting the non-linear responses of atmospheric systems to emission changes. This study underscores the importance of multi-faceted air quality management strategies that account for the intricate interplay of different emission sources and atmospheric processes. The insights gained from this unique period should inform the development of more effective, evidence-based air quality policies for a healthier and more sustainable future.

Dynamic thunderstorm capture and path imaging driven by information granulation and fuzzy C-means clustering

New semi-supervised fuzzy C-means clustering with asymmetric deviation constraints and fast algorithm

Rapid detection of petroleum residue properties using near-infrared spectroscopy integrated with fuzzy c-means clustering and competitive adaptive reweighted sampling variable selection.

Objective This study presented a comprehensive characterization of the osteomyelitis immune microenvironment, identified driver genes and pathogenic cell populations underlying disease progression, and uncovered potential therapeutic targets through single-cell and bulk transcriptomic analysis. Methods We analyzed time-series transcriptomic sequencing data from mouse osteomyelitis samples in the dataset GSE168896. Fuzzy c-means clustering was applied to reveal gene sets linked to disease progression. Immune cell infiltration analysis was conducted through the online tool ImmuCellAI-mouse. Furthermore, by leveraging single-cell sequencing data, we characterized immune cell subpopulations and pinpointed the key cell subtypes that were present in the osteomyelitis mice. Results We identified six gene clusters exhibiting distinct temporal expression patterns and functional roles in osteomyelitis, such as leukocyte and lymphocyte activation and ossification. Single-cell sequencing analysis further showed seven distinct cellular subpopulations. Among these, macrophages demonstrated a significant increase following osteomyelitis, and the infiltration of Mif + Cd63 + , Arg1 + Sdc4 + , and Cxcl1 + Ccl4 + macrophages significantly increased. Moreover, Ccl3–Ccr1 and Cxcl2–Cxcr2 ligand–receptors contributed mostly in immune cells. Conclusion Our findings tracked the transcriptional dynamics and evolving immune landscape of osteomyelitis, highlighting macrophages as central regulators of disease progression. We identified that significant infiltration of Arg1 + Sdc4 + , Cxcl1 + Ccl4 + , and Mif + Cd63 + macrophages may affect osteomyelitis through the Ccl3–Ccr1 and Cxcl2–Cxcr2 signaling pathways. These findings offer a new perspective on immune regulation in osteomyelitis.

Abstract One of the key elements influencing machining precision is the thermal inaccuracy of the CNC machine tool feeding mechanism. To improve prediction capability, this article presents a Transformer-GRU composite model for thermal error prediction, which adopts an information fusion method to consider the combined effects of multiple factors on thermal error. First, an experimental platform is constructed to measure thermal error, temperature and vibration data in the feeding system. Fuzzy C-means and Grey Relational Analysis are adopted to screen out screen out the data of temperature sensitive points, and vibration data are processed by means of Empirical Mode Decomposition with Wavelet Thresholding, Spearman Correlation Coefficient and Kernel Principal Component Analysis. Then, the temperature and vibration data are fused using multi-source information fusion technique. A Transformer-GRU composite model that embeds the Gated Recurrent Unit (GRU) into Transformer architecture is introduced. Finally, the temperature, vibration and displacement data are fused to construct a composite Transformer-GRU model for thermal error prediction of the feeding system. To prove the effectiveness of the presented model, its prediction results are compared and analyzed with Transformer and Transformer-LSTM models. The performance metrics show that the RMSE, MAE, and MSE of the Transformer-GRU composite model are 1.40, 1.15, and 1.97, respectively, which are 36.9%, 15.4%, and 29.9% lower than those of the Transformer's 1.68, 1.36, and 2.81, respectively, and those of the Transformer-LSTM's 2.22, 2.89, 4.91 reduced by 36.9%, 39.1%, 59.9%, respectively. The results demonstrate that the Transformer-GRU composite model, which utilizes multi-source information fusion, achieves higher prediction accuracy.

To address the issues of suboptimal sparsity and the tendency of clustering algorithms to converge to local optima in the estimation of the mixing matrix within underdetermined blind source separation (UBSS) systems, a novel mixing matrix estimation algorithm based on source signal sparsity is proposed. Firstly, the principle of underdetermined mixing matrix estimation leveraging source sparsity is derived. Building upon this foundation, improvements are made from enhancement of signal sparsity and optimization of the clustering algorithm. To overcome the limited sparse representation capability of conventional time-frequency (TF) transformation methods, a source signal sparsity enhancement algorithm based on the Local Maximum Synchroextracting Transform (LMSET) is proposed. This method rearranges the TF coefficients by detecting local maxima in the frequency direction, thereby achieving a more desirable TF resolution and enhanced signal sparsity. Furthermore, to mitigate the sensitivity of the Fuzzy C-Means (FCM) algorithm to initial cluster centers and its propensity for local optima, a robust FCM algorithm optimized by the PID(Proportional-integral-Derivative)-based Search Algorithm (PSA) is adopted. Simulation results demonstrate that the proposed algorithm achieves a superior TF representation and enhances the sparsity of source signals across various environments. Compared to traditional algorithms, the estimation accuracy of the mixing matrix is increased by 19.8%, effectively improving the performance of mixing matrix estimation in underdetermined blind source separation systems.

To address the challenges of real-time control in power systems with high renewable penetration, identifying historical transmission sections similar to future scenarios enables efficient reuse of mature control strategies. However, existing data-driven identification methods exhibit two primary limitations: they typically rely on static Total Transfer Capacity (TTC), ignoring the rapid regulation capability of Energy Storage Systems (ESS) in alleviating congestion; and they employ fixed weights for similarity measurement, failing to distinguish the varying importance of different features (e.g., critical line flows vs. ordinary voltages). To overcome these issues, this paper proposes a similarity identification method for transmission sections considering ESS regulation capabilities and adaptive feature weights. First, a hierarchical decision model is utilized to screen basic grid features. An optimization model incorporating ESS charge/discharge constraints and emergency power support potential is established to calculate the Dynamic TTC, constructing a multi-scale feature set that reflects the real-time safety margin of the grid. Second, a Dispersion-Weighted Fuzzy C-Means (DW-FCM) clustering algorithm is proposed. By introducing a dispersion-weighting mechanism, the algorithm utilizes data distribution characteristics to automatically learn and assign higher weights to key features with high distinguishability during the iteration process, overcoming the subjectivity of manual weighting. Furthermore, fuzzy validity indices (XB, PC, FS) are introduced to adaptively determine the optimal number of clusters. Finally, case studies on the IEEE 39-bus system verify that the proposed method significantly improves identification accuracy compared to traditional methods and provides more reliable references for dispatching decisions.

Nigeria faces persistent energy supply challenges, particularly in its northeastern region, where grid access is limited and dependence on fossil fuels undermines sustainability goals. Although the National Renewable Energy Action Plan (NREAP 2015–2030) outlines ambitious targets for renewable energy integration, it notably lacks specific strategies for geothermal development—leaving a critical gap in policy and resource utilization. This study addresses that gap by developing a scalable, cost-effective geothermal prospectivity mapping framework using remote sensing and aeromagnetic data integrated through a hybrid machine learning model. A novel combination of Deep Belief Networks (DBN) for feature extraction and Fuzzy C-Means (FCM) clustering for spatial classification was employed, with optimization achieved using three metaheuristic algorithms: Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA). Among these, the DBN-SA model achieved the best internal validity, with superior Silhouette Score, Davies–Bouldin Index, and cluster compactness, ensuring robust and interpretable prospectivity results. Key geothermal indicators—including land surface temperature, vegetation stress, Curie depth, heat flow, and magnetic source depth—were derived from Landsat and airborne magnetic datasets. The resulting map classifies the study area into low, moderate, and high geothermal potential zones, with validation supported by geological correlation and the presence of known thermal features like the Wikki Warm Spring. Approximately one-third of the area was identified as high-potential, particularly over basement terrains with high heat production and structural permeability. This approach offers both scientific insight and practical direction for decentralized, low-carbon energy deployment in northeastern Nigeria, aligning with broader national renewable energy goals and filling a crucial gap in geothermal resource planning.

Implantation failure is the most common cause of pregnancy failure and is a major limiting factor in assisted reproduction. Plasma microRNA (miRNA) expression profiles show dynamic changes among individuals with successful implantation. However, the trajectory differences in plasma miRNA expression between women with implantation failure and success, as well as the potential role of those miRNAs, remain unclear. This study included 84 women who underwent single frozen-thawed blastocyst transfer in a natural cycle. For each patient, longitudinal plasma samples across five time points throughout the peri-implantation period (day0/D3/D5/D7/D9) were collected and underwent miRNA sequencing. The failure group (n = 27) encountered complete implantation failure, while the success group (n = 57) achieved a live birth. Using trajectory analysis and fuzzy c-means clustering, we identified dynamically differentially expressed (DDE) miRNAs and their dynamic expression patterns (DEPs) in a screening set (n = 52) and validated findings in an independent validation set (n = 32). Clinical correlations, functional annotation, prediction model, and in vitro validation of prioritized miRNA-target interactions were systematically conducted. Twenty-four DDE miRNAs (FDR < 0.05) exhibited five temporal patterns: recession (R), growth (G), D3-trough (T), multimodal (M), and D5-trough (T2). The success group predominantly showed M-pattern miRNAs (62.5%), while failures demonstrated pattern transitions (M → T/T2) and exclusive T-pattern expression. Clinical relevance revealed eight DDE miRNAs associated with at least one clinical variable, with the majority being associated with estradiol/progesterone levels. Functional enrichment implicated Wnt/mTOR pathways in embryo implantation and decidualization. Six DDE miRNAs were successfully replicated in the validation set, while the support vector machine (SVM) model achieved area under the curve (AUC) values of 0.816 (D0) and 0.870 (D3). Additionally, the association between hsa-miR-214-3p and its target gene CTNNB1 was further confirmed in Ishikawa cells. Understanding the dynamic landscape changes of the miRNA transcriptome in individuals with implantation failure will help identify dynamic biomarkers from ovulation to the post-implantation stage, providing new insights into the pathological mechanisms of implantation failure and facilitating the research and development of new therapies in the clinical setting.

This study devises an innovative LiDAR point cloud down-sampling strategy that capitalizes on the properties of Fuzzy C Means (FCM) clustering membership functions in each dimension. Traditional down-sampling methods frequently encounter difficulties in striking a balance between computational efficiency and feature preservation, particularly for large-scale datasets. To tackle this issue, our approach breaks down the three-dimensional simplification problem into independent one-dimensional analyses. Specifically, FCM clustering is carried out separately on the X, Y, and Z coordinates to generate dimension-wise membership functions. These functions are then intelligently integrated to calculate comprehensive importance scores for each point, facilitating adaptive sampling that eliminates redundant data while retaining critical geometric features. Experimental results demonstrate that our method outperforms conventional approaches, including voxel grid, random, and farthest point sampling, in terms of geometric fidelity. The proposed method shows strong potential for real-time applications involving large-scale point clouds in fields such as autonomous driving, robotic navigation, and 3D reconstruction.

Descriptions of the vocal repertoires of primates are essential to understanding aspects of behaviour and social ecology, and are increasingly useful in applied conservation. Sportive lemurs (Lepilemur spp.) are highly vocal but otherwise cryptic, making them promising candidates for bioacoustics research and monitoring; however, vocal repertoires have been described for only a handful of the 25 species, and the question of gradation within and between call types remains largely unexplored. Here we describe the call types and gradedness of the vocal repertoire of the Critically Endangered Nosy Be sportive lemur (Lepilemur tymerlachsoni). We recorded wild sportive lemur vocalisations over a six-week period in early 2023. From the spectrograms of these recordings, we manually classified 14 distinct call types, distinguishable both aurally and visually, representing the largest known repertoire for the genus and among the largest reported for any nocturnal primate. Depending on the call type, calls were produced singly, in sequences and combinations, or in extended bouts. We then used an unsupervised machine learning technique, fuzzy c-means clustering, to objectively classify the repertoire and quantify its graded structure using 25 acoustic measurements extracted from the spectrograms. Fuzzy clustering identified two acoustic clusters, revealing gradation both within call types and between clusters. These patterns were visualised with Uniform Manifold Approximation and Projection ('UMAP') dimensionality reduction. We present representative spectrograms for each call type, and compare our results with the published repertoires of other sportive lemur species. Our study provides a foundation for further behavioural research and acoustic-based conservation of L. tymerlachsoni.

This study proposes the FCM-RF-SMOTE framework to resolve the issue of data imbalance in real-time freeway traffic state classification. The framework integrates Fuzzy C-Means (FCM), Random Forest (RF), and the Synthetic Minority Over-sampling Technique (SMOTE). Traffic states are classified into four categories (smooth, stable, congested, and severely congested) based on quantitative thresholds derived from FCM clustering centers. The validation utilizes SUMO simulation with Gaussian noise and a 10 Hz sampling rate to approximate millimeter-wave radar characteristics. Results show that the proposed framework significantly increases the representation of the severe congestion class from 3.67% to 19.83%. Consequently, the overall classification accuracy is enhanced from 77.67% to 97.80%, demonstrating superior performance in handling imbalanced datasets compared to baseline methods. The findings demonstrate the robustness of the algorithm for traffic monitoring systems, particularly in identifying minority traffic states, with future work planned for physical sensor validation.

General anesthesia comprises 3 essential components-hypnosis, analgesia, and immobility. Among these, maintaining an appropriate hypnotic state, or anesthetic depth, is crucial for patient safety. Excessively deep anesthesia may lead to hemodynamic instability and postoperative cognitive dysfunction, whereas inadequate anesthesia increases the risk of intraoperative awareness. Electroencephalography (EEG)-based monitoring has therefore become a cornerstone for evaluating anesthetic depth. However, processed electroencephalography (pEEG) indices remain vulnerable to various sources of interference, including electromyographic activity, interindividual variability, and anesthetic drug effects, which can yield inaccurate numerical outputs. With recent advances in machine learning, particularly unsupervised learning, data-driven methods that classify signals according to inherent patterns offer new possibilities for anesthetic depth analysis. This study aimed to establish a methodology for automatically identifying anesthesia depth using an unsupervised, machine learning-based clustering approach applied to pEEG data. Standard frontal EEG data from participants undergoing elective lumbar spine surgery were retrospectively analyzed, yielding more than 16,000 data points. The signals were filtered with a fourth-order Butterworth bandpass filter and transformed using the fast Fourier transform to estimate power spectral density. Normalized band power ratios for delta, high-theta, alpha, and beta frequencies were extracted as input features. Fuzzy C-Means (FCM) clustering (c=3, m=2) was applied to categorize anesthetic depth into slight, proper, and deep clusters. FCM clustering successfully identified 3 physiologically interpretable clusters consistent with EEG dynamics during progressive anesthesia. As anesthesia deepened, frontal alpha oscillations became more prominent within a delta-dominant background, while beta activity decreased with loss of consciousness. The fuzzy membership values quantified transitional states and captured the continuum of anesthetic depth. Visualization confirmed strong correspondence among cluster transitions, Patient State Index trends, and spectral density patterns. This study demonstrates the feasibility of using unsupervised machine learning to enhance anesthetic depth assessment. By applying FCM clustering to pEEG data, this approach improves the understanding of anesthesia depth and integrates effectively with existing monitoring modalities. The proposed FCM-based method complements current EEG indices and may assist anesthesia practitioners and even nonanesthesia professionals in assessing anesthetic depth to enhance patient safety.

While the energy levels of nodes in these networks vary, there is still a need for further advancements to fully optimize the performance of heterogeneous Wireless Sensor Networks (WSNs). Efficient data routing plays a vital role in improving the IoT-enabled WSNs overall performance. Existing routing protocols face difficulties in addressing issues like frequent node movement, optimizing energy efficiency, scalability, and adapting to changing network conditions. These challenges need to be addressed to ensure effective routing in IoT-based heterogeneous WSNs. Thus, a new approach for optimal routing in IoT-based heterogeneous WSNs that utilizes the STGOA is proposed. In the Network Modeling stage, various components of the IoT-enabled heterogeneous WSN, including sensor nodes, servers, the internet, BS, and the network topology, are defined. The interactions between these elements are established to define the communication structure and flow within the network. The Optimal Routing stage begins with the clustering of sensor nodes via the Fuzzy C Means (FCM) method. After clustering, the STGOA approach is employed to choose the optimal paths while considering critical constraints like energy consumption, trust, distance, security, and delay. The proposed STGOA method achieved maximum energy rate of 0.368 J at node 100 and 0.375 J at node 200 as compared to the other methods like GWO, SHO, GOA, STO, BOA, NBO, JFO, EEMCM and GMPSO.

This study proposes a fuzzy clustering-based vertical stacking strategy (FVSS) to reduce weight variance and enhance operational efficiency in container terminal operations. To address inefficiencies from reshuffling, defined as unnecessary container movements during retrieval, the FVSS method classifies containers into multiple weight classes using Fuzzy C-means (FCM) clustering based on historical container data. Stacking spaces are then proportionally allocated according to cluster sizes, and each stack is assigned a weight reference value to guide real-time container stacking. This enables flexible vertical stacking that dynamically adapts to weight similarities and uncertainty. The proposed strategy was evaluated against existing approaches including hybrid sequence stacking (HSS), random stacking strategy (RSS), and GMM-based methods. Numerical experiments using real-world terminal data demonstrate that FVSS outperforms other strategies, achieving up to 78% reduction in weight variance. Furthermore, performance remains stable even under uncertain weight conditions. These results highlight the practical advantage of integrating fuzzy optimization into stacking strategies, offering a robust and computationally efficient solution for container yard operations.

Identification of cognitive phenotypes in temporal lobe epilepsy and genetic generalized epilepsy using robotic assessment.

Reliable fault diagnosis in wind turbines relies on observational data such as that from the Supervisory Control and Data Acquisition (SCADA) system. However, the scarcity of labeled faults concerning such data requires the use of unsupervised anomaly detection algorithms (such as Fuzzy C-Means) to generate pseudo-labels indicating potential system faults. However, these pseudo-labels often lack transparency and reliability, confusing real fault signals with normal operational outliers. To overcome this challenge, this work proposes an innovative methodology that utilizes explainable artificial intelligence (XAI) techniques, specifically Shapley Additive Explanations (SHAP), and Local Interpretable Model-agnostic Explanations (LIME), to validate and refine pseudo-labels systematically. The methodology starts with anomaly identification and anomaly generation by unsupervised methods. Then, XAI is applied to explain which specific SCADA signals led to each anomaly classification. Finally, these explanations generated by XAI are analyzed from an engineering perspective to verify their correspondence with known patterns of mechanical or electrical failures. This step is crucial since it transforms the opaque pseudo-labels into “explained pseudo-labels”, which carry diagnostic information and allow the differentiation between genuine failures and false alarms. The fundamental contribution lies in the proactive use of XAI to validate the pseudolabel generation process, as opposed to its conventional use in interpreting predictive outputs. This application deepens the understanding of deviation signaling by the anomaly detection model, enabling the optimized selection of algorithms for specific types of failures and the identification of spurious normal operational outliers in wind turbines. It is expected that this work will provide a framework to enhance the reliability of unsupervised fault diagnosis, offering a clear understanding of anomaly origins and making alerts based on SCADA data more accurate. Additionally, the methodology enables the comparison of different algorithms, guiding their selection, and promoting a more interpretable approach to predictive maintenance in wind turbines.

Cluster-Based Insights into Cardiovascular and Autonomic Responses to Head-Up Tilt in Hypertension.

Damage mode identification for bonding interface of steel-UHPC composite structures based on acoustic emission and fuzzy C-means unsupervised clustering