Search Results (1,310)

Slope stability prediction is a critical task in geotechnical engineering, but machine learning (ML) models require large datasets, which are often costly and time-consuming to obtain. This study proposes a domain-driven teacher–student framework to overcome data limitations for predicting the dry factor of safety (FS dry). The teacher model, XGBoost, was trained on the original dataset to capture nonlinear relationships among key site-specific features (unit weight, cohesion, friction angle) and assign pseudo-labels to synthetic samples generated via domain-driven simulations. Six student models, random forest (RF), decision tree (DT), shallow artificial neural network (SNN), linear regression (LR), support vector regression (SVR), and K-nearest neighbors (KNN), were trained on the augmented dataset to approximate the teacher’s predictions. Models were evaluated using a train–test split and five-fold cross-validation. RF achieved the highest predictive accuracy, with an R² of up to 0.9663 and low error metrics (MAE = 0.0233, RMSE = 0.0531), outperforming other student models. Integrating domain knowledge and synthetic data improved prediction reliability despite limited experimental datasets. The framework provides a robust and interpretable tool for slope stability assessment, supporting infrastructure safety in regions with sparse geotechnical data. Future work will expand the dataset with additional field and laboratory tests to further improve model performance. Full article

Organisations are increasingly adopting the Chain of Custody (CoC) standards in the mining industry to enhance the traceability of minerals. It ensures that the minerals they have received are from credible sources and accompanied by verifiable information. However, unlikeother industries such as timber, where the effectiveness and benefits of CoC standards are mainly explored, this study subtly shifts the focus towards identifying strategic opportunities and innovation areas within the CoC standards that could extend beyond traceability. Four CoC standards were selected, and their provisions examined. It was found that implementing these requirements could not only enhance transparency but also support broader sustainability goals across the entire value chain. The study also identifies several challenges that could act as barriers to the CoC system, and these are seen as opportunities for innovative approaches to enhance the effectiveness of the standards. These are labelled as transformative innovation areas, and while they do include blockchains and analytical proof of origin technologies, this study also seeks to advocate for solutions that are more pragmatic and scalable. By identifying opportunities and areas of innovation, the findings will help improve the practical implementation of the standards and suggest areas for future evaluations of effectiveness that could consider aspects beyond traceability, such as sustainability and transparency. Full article

Spermatogenesis in teleosts is essential for reproductive function; however, it varies considerably among species. The testis of the viviparous molly fish (Poecilia sphenops) was examined using both ultrastructural and immunohistochemical methods. The testis displays a restricted lobular type, where germ cells develop synchronously within Sertoli cell-forming cysts. Transmission electron microscopy (TEM) revealed all stages of spermatogenesis. Mature sperm are at the apex of the cysts and migrate toward the sperm ducts. Sperm duct epithelium is lined by cuboidal cells joined by tight junctions, with apical cilia and desmosomal complexes contributing to transport and structural integrity. The sperm ducts showed strong Periodic Acid–Schiff (PAS)-positive expression among negative stained spermatocysts. Centrally, a cavity serves as a storage area for spermatozoa that are organized into unencapsulated bundles known as spermatozeugmata. Sertoli cells exhibited extended cytoplasmic processes that supported developing germ cells, whereas Leydig cells occupied the interstitial tissue, contributing to hormonal regulation. Immunohistochemical labeling demonstrated strong vimentin expression in Sertoli cells and telocytes, indicating their mesenchymal origin and structural role. Calretinin expression was confined to Leydig cells and certain ductal epithelial cells, supporting its use as a marker for steroidogenic and secretory functions. These findings provide new insights into the testicular specialization of P. sphenops, highlighting key somatic–germ cell interactions, ductal adaptations, and marker expression patterns that underlie male reproductive success in viviparous fish. Full article

Reinforcement Learning from Human Feedback (RLHF) produces powerful instruction-following models but relies on a preference-labeling process that is both costly and slow. An effective alternative, Reinforcement Learning from AI Feedback (RLAIF), uses large language models as teachers for relabeling; however, this introduces substantial label noise. In our setting, we found that AI teachers flipped approximately 50% of the original human preferences on the dataset, a condition that degrades the performance of standard direct preference optimization (DPO). We propose noise-robust DPO (nrDPO) and nrDPO-gated, two drop-in variants that make DPO resilient to noisy preferences. nrDPO reweights each pair by (i) a margin-confidence term from a frozen reference policy (base or SFT), (ii) a context-stability term that penalizes preferences that change under truncated histories, and (iii) a length correction to curb verbosity bias. nrDPO-gated further filters low-confidence pairs via a simple threshold on the reference margin. On a dataset with heavy synthetic noise (30% flips), nrDPO-gated improves the preference accuracy by +3.8% over vanilla DPO; in a realistic RLAIF setting, nrDPO-gated is the only configuration that recovers competitive alignment, reaching ≈60% on a 5k relabeled set (vs. ≈49–50% for vanilla DPO) and approaching RLHF baselines. Full article

Background: Segmenting nerve fibres in histological images is a tricky job because of how much the tissue looks can change. Modern neural network architectures, including U-Net and transformers, demonstrate varying degrees of effectiveness in this area. The aim of this study is to conduct a comparative analysis of the SegFormer, VGG-UNet, and FabE-Net models in terms of segmentation quality and speed. Methods: The training sample consisted of more than 75,000 pairs of images of different tissues (original slice and corresponding mask), scaled from 1024 × 1024 to 224 × 224 pixels to optimise computations. Three neural network architectures were used: the classic VGG-UNet, FabE-Net with attention and global context perception blocks, and the SegFormer transformer model. For an objective assessment of the quality of the models, expert validation was carried out with the participation of four independent pathologists, who evaluated the quality of segmentation according to specified criteria. Quality metrics (precision, recall, F1-score, accuracy) were calculated as averages based on the assessments of all experts, which made it possible to take into account variability in interpretation and increase the reliability of the results. Results: SegFormer achieved stable stabilisation of the loss function faster than the other models—by the 20–30th epoch, compared to 45–60 epochs for VGG-UNet and FabE-Net. Despite taking longer to train per epoch, SegFormer produced the best segmentation quality, with the following metrics: precision 0.84, recall 0.99, F1-score 0.91 and accuracy 0.89. It also annotated a complete histological section in the fastest time. Visual analysis revealed that, compared to other models, which tended to produce incomplete or excessive segmentation, SegFormer more accurately and completely highlights nerve structures. Conclusions: Using attention mechanisms in SegFormer compensates for morphological variability in tissues, resulting in faster and higher-quality segmentation. Image scaling does not impair training quality while significantly accelerating computational processes. These results confirm the potential of SegFormer for practical use in digital pathology, while also highlighting the need for high-precision, immunohistochemistry-informed labelling to improve segmentation accuracy. Full article

Intrinsically disordered regions (IDRs) are sometimes considered parts of the ‘dark proteomes’, i.e., protein parts that have been largely under-appreciated, as are the overlapping phenomena of low-complexity or compositionally biased regions (LCRs/CBRs). Experimentalists and computationalists alike are still learning how to decrypt the functionally meaningful features of such regions. Here, I report the creation of the support troupe Patterny to aid such protein cryptanalysis. The current troupe members are named Blocky, Bandy, Moduley, Repeaty, and Runny. To discern important features, protein regions are compared to ideal assortments wherein everything is sampled proportionally and dispersed randomly. Blocky discerns the segregation of amino-acids by type, and scores them for it. Bandy is focused on picking out compositional bands and calculating their evenness. Moduley labels the boundaries of optimized compositional modules (‘CModules’) and other possible boundary sets for compositionally biased regions. Repeaty concisely summarizes repetitiveness using an information entropy of amino-acid interval diversity. Runny enumerates homopeptide content and assesses its significance. Both original whole sequences and CModules from Moduley, are fed into the other Patterny members. Patterny is applied to some illustrative sample data from yeast proteome and the DISPROT database. It is available at Github, and might aid those aiming to intensify light-shedding and hypothesis generation for protein regions with function encoded in a distributed manner, such as IDRs and LCRs/CBRs more generally. Full article

Background/Objectives: Autism Spectrum Disorder (ASD) is a neurodevelopmental condition often accompanied by gastrointestinal (GI) symptoms and gut microbiota imbalances. The microbiota–gut–brain (MGB) axis is a bidirectional communication network linking gut microbes, the GI system, and the central nervous system (CNS). This narrative review explores the role of the MGB axis in ASD pathophysiology, focusing on communication pathways, neurodevelopmental implications, gut microbiota alteration, GI dysfunction, and emerging therapeutics. Methods: A narrative review methodology was employed. We searched major scientific databases including PubMed, Scopus, and Google Scholar for research on MGB axis mechanisms, gut microbiota composition in ASD, dysbiosis, leaky gut, immune activation, GI disorders, and intervention (probiotics, prebiotics, fecal microbiota transplantation (FMT), antibiotics and diet). Key findings from recent human, animal and in vitro studies were synthesized thematically, emphasizing mechanistic insights and therapeutic outcomes. Original references from the initial manuscript draft were retained and supplemented for comprehensiveness and accuracy. Results: The MGB axis involves neuroanatomical, neuroendocrine, immunological, and metabolic pathways that enable microbes to influence brain development and function. Individuals with ASD commonly exhibit gut dysbiosis characterized by reduced microbial diversity (notably lower Bifidobacterium and Firmicutes) and overpresentation of potentially pathogenic taxa (e.g., Clostridia, Desulfovibrio, Enterobacteriaceae). Dysbiosis is associated with increased intestinal permeability (“leaky gut”) and newly activated and altered microbial metabolite profiles, such as short-chain fatty acids (SCFAs) and lipopolysaccharides (LPSs). Functional gastrointestinal disorders (FGIDs) are prevalent in ASD, linking gut–brain axis dysfunction to behavioral severity. Therapeutically, probiotics and prebiotics can restore eubiosis, fortify the gut barrier, and reduce neuroinflammation, showing modest improvements in GI and behavioral symptoms. FMT and Microbiota Transfer Therapy (MTT) have yielded promising results in open label trials, improving GI function and some ASD behaviors. Antibiotic interventions (e.g., vancomycin) have been found to temporarily alleviate ASD symptoms associated with Clostridiales overgrowth, while nutritional strategies (high-fiber, gluten-free, or ketogenic diets) may modulate the microbiome and influence outcomes. Conclusions: Accumulating evidence implicates the MGB axis in ASD pathogenesis. Gut microbiota dysbiosis and the related GI pathology may exacerbate neurodevelopmental and behavioral symptoms via immune, endocrine and neural routes. Interventions targeting the gut ecosystem, through diet modification, probiotics, symbiotics, or microbiota transplants, offer therapeutic promise. However, heterogeneity in findings underscores the need for rigorous, large-scale studies to clarify causal relationships and evaluate long-term efficacy and safety. Understanding MGB axis mechanisms in ASD could pave the way for novel adjunctive treatments to improve the quality of life for individuals with ASD. Full article

Background: Electrocardiogram (ECG) signals are crucial for cardiovascular diagnosis, but their analysis face challenges from noise contamination, compression difficulties due to their non-stationary nature, and the inherent complexity of its morphological components, particularly for low-amplitude P- and T-waves obscured by noise. Methodology: This study proposes a novel, multi-stage framework for ECG signal denoising, compressing, and component decomposition. The proposed framework leverages the sparsity of ECG signal to denoise and compress these signals using autoencoder-driven dictionary (AE-DD) with matching pursuit. In this work, a data-driven dictionary was developed using a regularized autoencoder. Appropriate trained weights along with matching pursuit were used to compress the denoised ECG segments. This study explored different weight regularization techniques: L1- and L2-regularization. Results: The proposed framework achieves remarkable performance in simultaneous ECG denoising, compression, and morphological decomposition. The L1-DAE model delivers superior noise suppression (SNR improvement up to 18.6 dB at $-3$ dB input SNR) and near-lossless reconstruction ($\mathrm{MSE}<{10}^{-5}$). The L1-AE dictionary enables high-fidelity compression (CR = 28:1 ratio, $\mathrm{MSE}\approx 0.58\times {10}^{-5}$, PRD = 2.1%), outperforming non-regularized models and traditional dictionaries (DCT/wavelets), while its trained weights naturally decompose into interpretable sub-dictionaries for P-wave, QRS complex, and T-wave enabling precise, label-free analysis of ECG components. Moreover, the learned sub-dictionaries naturally decompose into interpretable P-wave, QRS complex, and T-wave components with high accuracy, yielding strong correlation with the original ECG ($r=0.98$, $r=0.99$, and $r=0.95$, respectively) and very low MSE ($1.93\times {10}^{-5}$, $9.26\times {10}^{-4}$, and $3.38\times {10}^{-4}$, respectively). Conclusions: This study introduces a novel autoencoder-driven framework that simultaneously performs ECG denoising, compression, and morphological decomposition. By leveraging L1-regularized autoencoders with matching pursuit, the method effectively enhances signal quality while enabling direct decomposition of ECG signals into clinically relevant components without additional processing. This unified approach offers significant potential for improving automated ECG analysis and facilitating efficient long-term cardiac monitoring. Full article

Intravital two-photon microscopy enables deep-tissue imaging of subcellular structures in live animals, but its original spatial resolution and image quality are limited by scattering, motion, and low signal-to-noise ratios. To address these challenges, we used a combination of tissue stabilization, denoising methods, and motion correction, together with resolution enhancement algorithms, including enhanced Super-Resolution Radial Fluctuations (eSRRF) and deconvolution, to acquire high-fidelity time-lapse images of internal organs. We applied this imaging pipeline to image genetically labeled mitochondria in vivo, in Dendra2 mice. Our results demonstrate that the eSRRF-combined method, compared to other evaluated algorithms, significantly shows improved spatial resolution and mitochondrial structure visualization, while each method exhibiting distinct strengths in terms of noise tolerance, edge preservation, and computational efficiency. These findings provide a practical framework for selecting enhancement strategies in intravital imaging studies targeting dynamic subcellular processes. Full article

Augmented reality (AR) technology offers a promising approach to providing consumers with detailed and personalized information about food products. The aim of this pilot study was to explore how the use of AR tools comprising visual and auditory formats affects consumers’ perception and understanding of nutrition labels of two commercially available products (lasagne ready meal and strawberry yogurt). The nutritional information of both the lasagne and yogurt product were presented to consumers (n = 30) under three experimental conditions: original packaging, visual AR, and visual and audio AR. Consumers answered questions about their perceptions of the products’ overall healthiness, caloric content, and macronutrient composition, as well as how the information was presented. The results showed that while nutritional information presented under the original packaging condition was more effective in changing consumer perceptions, the AR tools were found to be more “novel” and “memorable”. More specifically, for both lasagne and yogurt, the visual AR tool resulted in a more memorable experience compared to original packaging. The use of visual AR and visual and audio AR tools were considered novel experiences for both products. However, the provision of nutritional information had a greater impact on product perception than the specific experimental condition used to present it. These results provide evidence from a pilot study supporting the development of an AR tool for displaying and potentially improving the understanding of nutrition labels. Full article

Antibody function is an important topic for the understanding of disease, but it would be quite challenging to make an accurate prediction of a paratope position from very limited information such as a B cell receptor’s (BCR’s) amino acid sequence alone. In this context, this study presents a knowledge-based Bidirectional Encoder Representation from Transformers (K-BERT) to deliver a precise prediction of a paratope position from a B cell receptor’s amino acid sequence alone. Here, the knowledge context of an amino acid consisted of its predecessor amino acids. These knowledge contexts were either common among all amino acids within each BCR chain or different for different amino acids within each BCR chain. Also, oversampling was employed given that the original data of 20,679 cases (900 BCR chains) were characterized by class imbalance, i.e., 18,724:1955 for labels 0:1. The performance measures in terms of sensitivity and F1 registered great improvements as different knowledge contexts and oversampling were introduced. The accuracy, sensitivity, specificity and F1 of a baseline model (with common knowledge and no oversampling) were 90.3, 0.0, 100.0 and 50.0, respectively. On the other hand, the corresponding accuracy, sensitivity, specificity and F1 of the final model (with different knowledge and strong oversampling) were 83.2, 90.1, 78.1 and 84.1. The final model demonstrated better sensitivity and F1 outcomes compared to the baseline model, i.e., 90.1 vs. 0.0 for sensitivity, 84.1 vs. 50.0 for F1. In conclusion, the K-BERT is an effective decision support system to predict a paratope position from a B cell receptor’s amino acid sequence alone. It has great potential for antibody therapeutics. Full article

The study of horse genetic diversity is imperative for informing conservation strategies, safeguarding ancestral lineages, and enhancing breed adaptability to environmental and disease pressures. This study employed 13 microsatellite markers with fluorescent-labeled capillary electrophoresis to analyze the genetic diversity of the Kyrgyz horse (n = 30) and Barkol horse (n = 30) for the first time, comparing them with three other indigenous horse breeds (n = 30 per breed) from Xinjiang, China. A total of 208 alleles were detected. The Polymorphic Information Content (PIC) results from GenAlEx 6.5115 show that all loci, except for the HTG06 locus in the Yanqi horse races, were highly polymorphic (PIC > 0.5), indicating a high level of genetic diversity across the five races. Among the five races, the Kyrgyz horse exhibited the lowest mean values for the effective number of alleles (Ne), observed heterozygosity (Ho), and expected heterozygosity (He), which were 6.025, 0.737, and 0.810, respectively. In contrast, the Barkol horse showed the highest mean number of alleles (Na), Ne, and He values, at 11.308, 6.330, and 0.816, respectively. Principal Coordinate Analysis (PCoA), performed using GenAlEx 6.5115, revealed the smallest genetic distance between the Kyrgyz and Yanqi horse breeds. Combined with phylogenetic tree and clustering analysis results, this supports the hypothesis that the two breeds share a common origin. This study offers valuable scientific insights for conserving and utilizing the genetic resources of indigenous Xinjiang horse breeds, specifically the Kyrgyz and Barkol horses. Full article

Graph Neural Networks (GNNs) have become a cornerstone for graph classification, yet their vulnerability to backdoor attacks remains a significant security concern. While clean-label attacks provide a stealthier approach by preserving original labels, they tend to be less effective in graph settings compared to traditional dirty-label methods. This performance gap arises from the inherent dominance of rich, benign structural patterns in target-class graphs, which overshadow the injected backdoor trigger during the GNNs’ learning process. We demonstrate that prior strategies, such as adversarial perturbations used in other domains to suppress benign features, fail in graph settings due to the amplification effects of the GNNs’ message-passing mechanism. To address this issue, we propose two strategies aimed at enabling the model to better learn backdoor features. First, we introduce a long-distance trigger injection method, placing trigger nodes at topologically distant locations. This enhances the global propagation of the backdoor signal while interfering with the aggregation of native substructures. Second, we propose a vulnerability-aware sample selection method, which identifies graphs that contribute more to the success of the backdoor attack based on low model confidence or frequent forgetting events. We conduct extensive experiments on benchmark datasets such as NCI1, NCI109, Mutagenicity, and ENZYMES, demonstrating that our approach significantly improves attack success rates (ASRs) while maintaining a low clean accuracy drop (CAD) compared to existing methods. This work offers valuable insights into manipulating the competition between benign and backdoor features in graph-structured data. Full article

Although neural network-based methods have significantly advanced the field of machine fault diagnosis, they remain vulnerable to physical adversarial attacks. This work investigates such attacks in the physical context of a real production line. Attacks simulate failures or irregularities arising from the maintenance or production department during the production process, a scenario commonly encountered in industrial environments. The experiments are conducted using data from vibration signals and operational parameters of a motor installed in an industrial air-cooling system used for staple fiber production. In this context, we propose the Mean Confusion Impact Index (MCII), a novel and simple robustness metric that measures the average misclassification confidence of models under adversarial physical attacks. By performing a series of hardware-level interventions, this work aims to demonstrate that even minor physical disturbances can lead to a significant reduction in the model’s diagnostic accuracy. Additionally, a hybrid defense approach is proposed, which leverages deep feature representations extracted from the original classification model and integrates them with lightweight classifiers retrained on adversarial labeled data. Research findings underscore an important limitation in existing industrial artificial intelligence (AI)-based monitoring systems and introduce a practical, scalable framework for improving the physical resilience of machine fault diagnosis in real-world environments. Full article

Chest X-ray (CXR) imaging is essential for diagnosing thoracic diseases, and computer-aided diagnosis (CAD) systems have made substantial progress in automating the interpretation of CXR images. However, some existing methods often overemphasize local features while neglecting global context, limiting their ability to capture the broader pathological landscape. Moreover, most methods fail to model label correlations, leading to insufficient utilization of prior knowledge. To address these limitations, we propose a novel multi-label CXR image classification framework, termed the Label Masking-enhanced Residual Attention Network (LMeRAN). Specifically, LMeRAN introduces an original label-specific residual attention to capture disease-relevant information effectively. By integrating multi-head self-attention with average pooling, the model dynamically assigns higher weights to critical lesion areas while retaining global contextual features. In addition, LMeRAN employs a label mask training strategy, enabling the model to learn complex label dependencies from partially available label information. Experiments conducted on the large-scale public dataset ChestX-ray14 demonstrate that LMeRAN achieves the highest mean AUC value of 0.825, resulting in an increase of 3.1% to 8.0% over several advanced baselines. To enhance interpretability, we also visualize the lesion regions relied upon by the model for classification, providing clearer insights into the model’s decision-making process. Full article