Search Results (6,421)

Decoding visual content from neural activity remains a central challenge at the intersections of engineering, neuroscience, and computational modeling. Prior work has primarily leveraged electroencephalography (EEG) with generative models to recover static images. In this study, we advance EEG-based decoding by introducing a temporal encoding framework that approximates dynamic object transformations across time. EEG recordings from healthy participants (n = 20) were used to model neural representations of objects presented in “initial” and “later” states. Individualized classifiers trained on time-specific EEG signatures achieved high discriminability, with Random Forest models reaching a mean accuracy and standard deviation of 92 ± 2% and a mean AUC-ROC and standard deviation of 0.87 ± 0.10, driven largely by gamma- and beta-band activity at the frontal electrodes. These results confirm and extend evidence of strong interindividual variability, showing that subject-specific models outperform intersubject approaches in decoding temporally varying object representations. Beyond classification, we demonstrate that pairwise temporal encodings can be integrated into a generative pipeline to produce approximated reconstructions of short video sequences and 3D object renderings. Our findings establish that temporal EEG features, captured using low-cost open-source hardware, are sufficient to support the decoding of visual content across discrete time points, providing a versatile platform for potential applications in neural decoding, immersive media, and human–computer interaction. Full article

Memory forensics is an essential cybersecurity tool that comprehensively examines volatile memory to detect the malicious activity of fileless malware that can bypass disk analysis. Image-based detection techniques provide a promising solution by visualizing memory data into images to be used and analyzed by image processing tools and machine learning methods. However, the effectiveness of image-based data for detection and classification requires high computational efforts. This paper investigates the efficacy of texture-based methods in detecting and classifying memory-resident or fileless malware using different image resolutions, identifying the best feature descriptors, classifiers, and resolutions that accurately classify malware into specific families and differentiate them from benign software. Moreover, this paper uses both local and global descriptors, where local descriptors include Oriented FAST and Rotated BRIEF (ORB), Scale-Invariant Feature Transform (SIFT), and Histogram of Oriented Gradients (HOG) and global descriptors include Discrete Wavelet Transform (DWT), GIST, and Gray Level Co-occurrence Matrix (GLCM). The results indicate that as image resolution increases, most feature descriptors yield more discriminative features but require higher computational efforts in terms of time and processing resources. To address this challenge, this paper proposes a novel approach that integrates Local Interpretable Model-agnostic Explanations (LIME) with deep learning models to automatically identify and crop the most important regions of memory images. The LIME’s ROI was extracted based on ResNet50 and MobileNet models’ predictions separately, the images were resized to 128 × 128, and the sampling process was performed dynamically to speed up LIME computation. The ROIs of the images are cropped to new images with sizes of (100 × 100) in two stages: the coarse stage and the fine stage. The two generated LIME-based cropped images using ResNet50 and MobileNet are fed to the lightweight neural network to evaluate the effectiveness of the LIME-based identified regions. The results demonstrate that the LIME-based MobileNet model’s prediction improves the efficiency of the model by preserving important features with a classification accuracy of 85% on multi-class classification. Full article

The objective of this study is to analyze spatial entropy flows that reveal the directional dynamics of climate change—patterns that remain obscured in traditional statistical analyses. This approach enables the identification of pathways for “climate information transport”, highlights associations with atmospheric circulation types, and allows for the localization of both sources and “informational voids”—regions where entropy is dissipated. The analytical framework is grounded in a quantitative assessment of long-term climate variability across Europe over the period 1901–2010, utilizing Shannon entropy as a measure of atmospheric system uncertainty and variability. The underlying assumption is that the variability of temperature and precipitation reflects the inherently dynamic character of climate as a nonlinear system prone to fluctuations. The study focuses on calculating entropy estimated within a 70-year moving window for each calendar month, using bivariate distributions of temperature and precipitation modeled with copula functions. Marginal distributions were selected based on the Akaike Information Criterion (AIC). To improve the accuracy of the estimation, a block bootstrap resampling technique was applied, along with numerical integration to compute the Shannon entropy values at each of the 4165 grid points with a spatial resolution of 0.5° × 0.5°. The results indicate that entropy and its derivative are complementary indicators of atmospheric system instability—entropy proving effective in long-term diagnostics, while its derivative provides insight into the short-term forecasting of abrupt changes. A lag analysis and Spearman rank correlation between entropy values and their potential supported the investigation of how circulation variability influences the occurrence of extreme precipitation events. Particularly noteworthy is the temporal derivative of entropy, which revealed strong nonlinear relationships between local dynamic conditions and climatic extremes. A spatial analysis of the information entropy field was also conducted, revealing distinct structures with varying degrees of climatic complexity on a continental scale. This field appears to be clearly structured, reflecting not only the directional patterns of change but also the potential sources of meteorological fluctuations. A field-theory-based spatial classification allows for the identification of transitional regions—areas with heightened susceptibility to shifts in local dynamics—as well as entropy source and sink regions. The study is embedded within the Fokker–Planck formalism, wherein the change in the stochastic distribution characterizes the rate of entropy production. In this context, regions of positive divergence are interpreted as active generators of variability, while sink regions function as stabilizing zones that dampen fluctuations. Full article

The desire to improve the safety of navigation, especially in restricted and very busy areas like the straits, leads researchers to evaluate possible uses of Artificial Intelligence as an alternative to the traditional probabilistic methods. This is possible thanks to the large amount of available AIS data generated by ships in transit. In this work, a Machine Learning algorithm (Classification Decision Tree) was trained with eight features coming from AIS data of the Strait of Messina (Italy), with the aim of carrying out a two-class classification of the single AIS data to find anomalies in ship transits that could compromise navigation safety. Since anomalous events are relatively rare, compared to the large amount of information related to the normal navigation situations, the challenge of this work was to obtain an artificial dataset with the aim of simulating the possible anomalous navigation conditions for the Strait investigated, known the active risk mitigation means one. For this reason, the dataset containing abnormal events was obtained simulating different risk scenarios. A hyperparameters tuning with a Bayesian optimization approach and a 5-fold cross validation have been performed to improve the quality of the model and a large dataset has been tested. The accuracy of both validation and test phases is <99.5% and <95.9%, respectively. This can make it possible to identify anomalous navigation conditions in real time, in order to quickly classify possible conditions of risk. The method can be used as a Decision Support Tool by the authority in order to improve the capacity of the single operator to identify the possible risk situation inside the Strait of Messina. Full article

Honey is a natural product of high nutritional and therapeutic value, whose biological properties are closely linked to its botanical origin and chemical composition. This study aimed to characterize avocado honey in terms of botanical origin, physicochemical parameters, phenolic content, antioxidant activity, chemical profile by LC-MS/MS, and antibacterial potential. Melissopalynological analysis revealed 86.21% avocado pollen, allowing classification as monofloral honey. The sample presented amber color and a high total phenolic content (269.79 ± 1.10 mg GAE 100 g⁻¹), values higher than those commonly reported for Brazilian and international honeys. Antioxidant activity, assessed by the DPPH method, confirmed the strong radical-scavenging capacity, consistent with the phenolic profile identified (EC₅₀ 10.250 ± 0.003 mg mL⁻¹). LC-MS/MS analysis allowed the annotation of nine compounds, including caffeine, scopoletin, abscisic acid, and vomifoliol, compounds associated with antioxidant, anti-inflammatory, and metabolic regulatory activities. Although no antibacterial effect was detected against the tested oral bacterial strains, the results highlight the chemical diversity and functional potential of avocado honey. Overall, the findings reinforce the bioactive potential of avocado honey, particularly due to its strong antioxidant capacity and diversity of metabolites, supporting its value as a natural resource of nutritional and therapeutic interest. Full article

Seagrass meadows are vital for biodiversity and provide a plethora of ecosystem services, but significant losses due to human activity and climate change have been observed in recent decades. This study aims to evaluate whether the integration of Sentinel-2 composites, cloud computing (Google Earth Engine, GEE), and machine learning (ML) classifiers can produce accurate, scalable maps of seagrass habitats, enabling reliable estimates of associated carbon stocks. In this case study, we developed a methodological workflow for local-scale seagrass mapping in Cyprus, covering a total area of 310 km². ML techniques, specifically the Random Forest (RF) classifier and Classification And Regression Tree (CART), were employed in the main processing stage. The RF classifier achieved an overall accuracy of 73.5%, with a seagrass-specific F1-score of 69.4%. Class-specific F1-scores ranged from 63.2% for hard bottoms to 98.2% for deep water, accounting for variability in habitat separability. The workflow is designed to be scalable across Cyprus and potentially the broader EMMENA region (Eastern Mediterranean, Middle East, and North Africa). Based on the mapped extent of Posidonia oceanica meadows, preliminary estimates suggest a carbon stock of approximately 19,000 Mg C in Cyprus. Full article

Background/Objective: Biochemical recurrence (BCR) after radical prostatectomy (RP) for prostate cancer indicates disease progression. Although type 2 diabetes mellitus (T2D) shows a paradoxical association with prostate cancer risk, the prognostic role of T2D-related genetic variants remains unclear. Methods: We analyzed 113 common T2D susceptibility-related single-nucleotide polymorphisms (SNPs) in 644 Taiwanese men with localized prostate cancer (D’Amico risk classification: 12% low, 34% intermediate, and 54% high) treated with RP. Associations between SNPs and BCR were assessed using Cox regression, adjusting for key clinicopathological factors. Functional annotation was performed using HaploReg and FIVEx, while The Cancer Genome Atlas transcriptomic data were analyzed for C2 calcium-dependent domain-containing 4A (C2CD4A) expression. Gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were applied to explore related biological pathways. Results: C2CD4A SNP rs4502156 was independently associated with a reduced risk of BCR (hazard ratio = 0.80, p = 0.035). The protective C allele correlated with higher C2CD4A expression. Low C2CD4A expression is associated with advanced pathological stages, higher Gleason scores, and disease progression. GSEA revealed negative enrichment of mitotic and chromatid segregation pathways in high-C2CD4A-expressing tumors, with E2F targets being the most suppressed. GSVA confirmed an inverse correlation between C2CD4A expression and E2F pathway activity, with CDKN2C as a co-expressed functional gene. Conclusions: The T2D-related variant rs4502156 in C2CD4A independently predicts a lower risk of BCR, potentially via suppression of the E2F pathway, and may serve as a germline biomarker for postoperative risk stratification. Full article

In pursuit of sustainable development goal 5 (SDG5), this study underscores gender equity and women’s empowerment as pivotal themes in sustainable development. It examines the drivers of women’s empowerment, including education, economics, finance, and legal rights, using data from $n=223$ individuals, primarily women (68.4%) aged 20–30 (69.6%). The research methodology integrates descriptive statistical measures, machine learning (ML) algorithms, and graphical representations to systematically explore the fundamental research inquiries that align with SDG5, which focuses on achieving gender equity. The results indicate that higher educational levels, captured through ordinal encoding and correlation analyzes, are strongly linked to increased labor market participation and entrepreneurial activity. The random forest (RF) and support vector machine (SVM) classifiers achieved overall accuracies of 89% and 93% for the categorization of experience, respectively. Although 91% of women have bank accounts, only 47% reported financial independence due to gendered barriers. Logistic regression correctly identified financially independent women with a 93% recall, but the classification of non-independent participants was less robust, with a 44% recall. Access to legal services, modeled using a neural network, was a potent predictor of empowerment (F1-score 0.83 for full access cases), yet significant obstacles persist for those uncertain about or lacking legal access. These findings underscore that, while formal institutional access is relatively widespread among educated women literate in the digital world, perceived and practical barriers in the financial and legal realms continue to hinder empowerment. The results quantify these effects and highlight opportunities for tailored, data-driven policy interventions targeting persistent gaps. Full article

Lameness remains a significant welfare and productivity challenge in dairy farming, often underdiagnosed due to the limitations of conventional detection methods. Unlike most previous approaches to lameness detection that rely on a single-sensor or gait-based measurement, this study integrates four complementary data domains—behavioural, physiological, biochemical, and milk composition parameters—collected from 272 dairy cows during early lactation to enhance diagnostic accuracy and biological interpretability. The main objective of this study was to evaluate and compare the diagnostic classification performance of multiple machine learning (ML) algorithms trained on multimodal data collected at the time of clinical lameness diagnosis during early lactation, and to identify the most influential physiological and biochemical traits contributing to classification accuracy. Specifically, six algorithms—random forest (RF), neural network (NN), Ensemble, support vector machine (SVM), k-nearest neighbors (KNN), and logistic regression (LR)—were assessed. The input dataset integrated physiological parameters (e.g., water intake, body temperature), behavioural indicators (rumination time, activity), blood biochemical biomarkers (non-esterified fatty acids (NEFA), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), gamma-glutamyl transferase (GGT)), and milk quality traits (fat, protein, lactose, temperature). Among all models, RF achieved the highest validation accuracy (97.04%), perfect validation specificity (100%), and the highest normalized Matthews correlation coefficient (nMCC = 0.94), as determined through Monte Carlo cross-validation on independent validation sets. Lame cows showed significantly elevated NEFA and body temperatures, reflecting enhanced lipid mobilization and inflammatory stress, alongside reduced water intake, milk protein, and lactose content, indicative of systemic energy imbalance and impaired mammary function. These physiological and biochemical deviations emphasize the multifactorial nature of lameness. Linear models like LR underperformed, likely due to their inability to capture the non-linear and interactive relationships among physiological, biochemical, and milk composition features, which were better represented by tree-based and neural models. Overall, the study demonstrates that combining sensor data with blood biomarkers and milk traits using advanced ML models provides a powerful, objective tool for the clinical classification of lameness, offering practical applications for precision livestock management by supporting early, data-driven decision-making to improve welfare and productivity on dairy farms. Full article

Background/Objectives: Ovarian cancer is a heterogeneous malignancy with molecular subtypes that strongly influence prognosis and therapy. High-dimensional mRNA data can capture this biological diversity, but its complexity and noise limit robust subtype characterization. Furthermore, current classification approaches often fail to reflect subtype-specific transcriptional programs, underscoring the need for computational strategies that reduce dimensionality and identify discriminative molecular features. Methods: We designed a multi-stage feature selection and network analysis framework tailored for high-dimensional transcriptomic data. Starting with ~65,000 mRNA features, we applied unsupervised variance-based filtering and correlation pruning to eliminate low-information genes and reduce redundancy. The applied supervised Select-K Best filtering further refined the feature space. To enhance robustness, we implemented a hybrid selection strategy combining recursive feature elimination (RFE) with random forests and LASSO regression to identify discriminative mRNA features. Finally, these features were then used to construct a gene co-expression similarity network. Results: This pipeline reduced approximately 65,000 gene features to a subset of 83 discriminative transcripts, which were then used for network construction to reveal subtype-specific biology. The analysis identified four distinct groups. One group exhibited classical high-grade serous features defined by TP53 mutations and homologous recombination deficiency, while another was enriched for PI3K/AKT and ARID1A-associated signaling consistent with clear cell and endometrioid-like biology. A third group displayed drug resistance-associated transcriptional programs with receptor tyrosine kinase activation, and the fourth demonstrated a hybrid profile bridging serous and endometrioid expression modules. Conclusions: This pilot study shows that combining unsupervised and supervised feature selection with network modeling enables robust stratification of ovarian cancer subtypes. Full article

Increasing macroeconomic uncertainty necessitates that firms optimize their R&D investment and commercialization strategies. Patents, as crucial outcomes of R&D with legal protection, impose significant costs due to progressively increasing maintenance fees. Predicting patent life accurately thus becomes critical for effective patent management. Previous studies have often and primarily employed classification models for patent life prediction, while limiting practical utility due to coarse granularity. This study proposes a robust ensemble regression model combining multiple machine learning techniques, such as Random Forest and deep neural networks, to directly predict patent life. The proposed model achieved superior performance, surpassing individual baseline models, and recorded a Mean Absolute Error (MAE) of approximately 852.81. Additional validation with active patents further demonstrated the model’s practical feasibility, showing its potential to support sustainable intellectual property management by accurately predicting longer life for high-quality patents currently maintained. Consequently, the proposed model provides ongoing firms and brand-new startups with a decision support tool for strategic patent maintenance and commercialization decisions. By promoting efficient allocation of R&D resources and reducing unnecessary maintenance of low-value patents, the approach fosters sustainable management of innovation assets, enhancing predictive accuracy and long-term applicability. Full article

Keratin is a fibrous structural protein found in various natural materials such as hair, feathers, and nails. Its high stability and cross-linked structure make it resistant to degradation by common proteases, leading to the accumulation of keratinous waste in various industries. In this study, we developed and validated an effective bioinformatics-driven strategy for mining novel keratinase genes from the Esmatlas (ESM Metagenomic Atlas) macrogenomic database. Two candidate genes, ker820 and ker907, were identified through sequence alignment, structural modeling, and phylogenetic analysis, and were subsequently heterologously expressed in Escherichia coli Rosetta (DE3) with the assistance of a solubility-enhancing chaperone system. Both enzymes belong to the Peptidase S8 family. Enzymatic characterization revealed that GST-tagged ker820 and ker907 exhibited strong keratinolytic activity, with optimal conditions at pH 9.0 and temperatures of 60 °C and 50 °C, respectively. Both enzymes showed significant degradation of feather and cat-hair keratin. Kinetic analysis showed favorable catalytic parameters, including $K_m$ values of 9.81 mg/mL (ker820) and 5.25 mg/mL (ker907), and $V_{max}$ values of 120.99 U/mg (ker820) and 89.52 U/mg (ker907). Stability tests indicated that GST-ker820 retained 70% activity at 60 °C for 120 min, while both enzymes remained stable at 4 °C for up to 10 days. These results demonstrate the high catalytic capacity, thermal stability, and substrate specificity of the enzymes, supporting their classification as active keratinases. This study introduces a promising strategy for efficiently discovering novel functional enzymes using an integrated computational and experimental approach. Beyond keratinases, this methodology could be extended to screen for enzymes with potential applications in environmental remediation. Full article

Maternal tobacco smoke exposure is associated with impaired fetal growth and long-term disease risk (DOHaD, Developmental Origins of Health and Disease). Whether placental steroid hormones are independently altered remains a matter of debate. We quantified six placental steroids (estradiol, estriol, estrone, progesterone, testosterone, and pregnanediol) using HPLC–Corona CAD in 70 deliveries (C = 30; PS = 20; AS = 20). Distributional differences were assessed with Kruskal–Wallis and pairwise Mann–Whitney tests with Benjamini–Hochberg (BH) control. Adjusted associations used log-linear OLS with HC3 robust SE: Model A (gestational age, maternal BMI, newborn sex) and Model B (Model A + birth weight), reported as percent change vs. controls, computed as (exp(β) − 1) × 100 with 95% CI. Secondary analyses tested (i) multiclass logistic classification of C/PS/AS from the steroid panel (5-fold stratified CV) and (ii) prediction of birth weight (OLS and 2-component PLS). All six steroids differed by group (BH-adjusted p ranging from 9.18 × 10⁻¹² to 6.66 × 10⁻⁸). In Model A, AS vs. C showed lower estrogens/progestins (estradiol, −46.2%; estriol, −24.7%; estrone, −25.9%; progesterone, −28.2%; pregnanediol, −31.4%) and higher testosterone (+40.8%); these effects persisted in Model B after adjusting for birth weight. The panel classified C/PS/AS with 0.900 cross-validated accuracy (weighted OvR AUC 0.994). Hormones poorly predicted birth weight (PLS CV R² = −0.777). Maternal active and passive smoking is associated with a coherent and independent disruption of placental steroidogenesis. A targeted placental steroid panel offers biologically meaningful early markers relevant to DOHaD. Full article

This study investigates the impact of audio feedback on cognitive performance during VR puzzle games using EEG analysis. Thirty participants played three different VR puzzle games under two conditions (with and without audio) while their brain activity was recorded. To analyze concentration levels and neural engagement patterns, we employed spectral analysis combined with a preprocessing algorithm and an optimized Deep Neural Network (DNN) model. The proposed processing stage integrates feature normalization, automatic labeling based on Principal Component Analysis (PCA), and Gamma band feature extraction, transforming concentration detection into a supervised classification problem. Experimental validation was conducted under the two gaming conditions in order to evaluate the impact of multisensory stimulation on model performance. The results show that the proposed approach significantly outperforms traditional machine learning classifiers (SVM, LR) and baseline deep learning models (DNN, DGCNN), achieving a 97% accuracy in the audio scenario and 83% without audio. These findings confirm that auditory stimulation reinforces neural coherence and improves the discriminability of EEG patterns, while the proposed method maintains a robust performance under less stimulating conditions. Full article

Arable field abandonment is a major driver of landscape change in rural areas worldwide. It is defined as the cessation of agricultural activities and the withdrawal of agricultural management on land. This study examined arable land abandonment and subsequent land use and land cover (LULC) changes in Gotyibeni, Manqorholweni, Mawane, and Melani villages over a 20-year period. The aim was to understand these changes and how rural livelihoods and social relationships within and between households were perceived to have transformed following the LULC shifts. Landsat 5, 7, 8, and 9 multispectral imageries with a 30 m spatial resolution were analysed for two periods (i.e., 2000–2010 and 2010–2020). Five land cover classes were mapped: arable fields, grasslands, homestead gardens, residential areas, and shrublands. Post-classification change detection revealed a steady decline in arable fields, largely replaced by grasslands, shrublands, and residential areas. User accuracy was above 80% across all LULC maps, providing confidence in the LULC results. To link these spatial changes with social outcomes, 97 households that had abandoned field cultivation were purposively selected across the four villages. Semi-structured interviews were conducted to capture household experiences. Findings showed that reduced field cultivation was perceived to undermine household economic status, with households increasingly dependent on government social grants amid high unemployment. In addition, weakened social connections and shifts in cultural practices were reported. Overall, the study demonstrated how combining satellite imagery with community perspectives provides a comprehensive understanding of rural arable land abandonment and its consequences. Full article