Search Results (34,080)

Handling class imbalance remains a central challenge in machine learning, particularly in pattern recognition tasks where identifying rare but critical anomalies is of paramount importance. Traditional generative models often decouple data synthesis from classification, leading to a distribution mismatch that limits their practical benefit. To address these shortcomings, we introduce Causal Cooperative Networks (CCNETS), a modular framework that establishes a functional causal link between generation, inference, and reconstruction. CCNETS is composed of three specialized cooperative modules: an Explainer for latent feature abstraction, a Reasoner for probabilistic label prediction, and a Producer for context-aware data synthesis. These components interact through a dynamic causal feedback loop, where classification outcomes directly guide targeted sample synthesis to adaptively reinforce vulnerable decision boundaries. A key innovation, our proposed Zoint mechanism, enables the adaptive fusion of latent and observable features, enhancing semantic richness and decision-making robustness under uncertainty. We evaluated CCNETS on two distinct real-world datasets: Credit Card Fraud Detection dataset, characterized by extreme imbalance (fraud rate < 0.2%), and the AI4I 2020 Predictive Maintenance dataset (failure rate < 4%). Across comprehensive experimental setups, CCNETS consistently outperformed baseline methods, achieving superior F1-scores, and AUPRC. Furthermore, data synthesized by CCNETS demonstrated enhanced generalization and learning stability under limited data conditions. These results establish CCNETS as a scalable, interpretable, and hybrid soft computing framework that effectively aligns synthetic data with classifier objectives, advancing robust imbalanced learning. Full article

Background/Objective: Breast cancer (BC) management has traditionally relied on static clinicopathologic and immunohistochemical biomarkers (hormone receptor status, HER2 expression, and proliferative activity assessed at diagnosis). However, these biomarkers are typically evaluated at a single time point and may not reflect therapy-induced molecular evolution. This study evaluates whether longitudinal molecular profiling before and after treatment better characterizes tumor dynamics and provides clinically actionable insights into treatment response, resistance, and prognosis. Methods: Thirty-two patients with invasive breast carcinoma were analyzed using histopathology, immunohistochemistry, tissue-based next-generation sequencing, and plasma circulating tumor DNA (ctDNA) analysis. Paired tumor tissue and plasma samples were collected before and after treatment when available. Changes in biomarker expression, molecular subtype, and genomic alterations were assessed to characterize molecular plasticity under therapeutic pressure. Results: The cohort had a median age of 54 years (range 29–86), predominantly invasive ductal carcinoma (>85%), and high-grade disease. Hormone receptor-positive tumors accounted for 78.1%. Molecular subtypes were Luminal A (34.4%), Luminal B HER2− (40.6%), Luminal B HER2+ (6.3%), HER2-enriched (6.3%), and triple-negative breast cancer (12.5%). Initial tissue sequencing identified PI3K/AKT pathway alterations in 28.1% of cases. Post-treatment analyses revealed substantial molecular discordance, including progesterone receptor loss (33.3%), HER2 status changes (33.3%), and Ki67 variability (77.8%). Plasma ctDNA analysis was informative in 53.1% of patients and identified additional clinically relevant alterations, including FGFR1 amplification and BRCA1/2 variants not detected in tissue. Conclusions: BC molecular profiles are dynamic and frequently altered by therapy. Longitudinal molecular assessment reveals clinically actionable changes overlooked by static subtyping, supporting a dynamic model of molecular classification, highlighting the potential value of adaptive molecular subtyping to improve treatment stratification and resistance monitoring. Full article

This study aimed to explore genetic variants associated with serum albumin (ALB) levels in Chinese winter sports athletes using genome-wide association analysis (GWAS) and to investigate potential regulatory mechanisms using bioinformatics annotation. A total of 382 Chinese winter sports athletes were recruited. ALB levels were compared between elite and non-elite athletes. GWAS was conducted using PLINK v1.9, with ALB as the phenotype and sex, age, and principal components as covariates. Associated SNPs were annotated using GTEx and SNPnexus. No significant differences were observed in ALB levels between elite and non-elite male or female athletes, and ALB levels in all groups followed a normal distribution. We identified 113 SNPs reaching a suggestive significance threshold (p < 1 × 10⁻⁵), with per-variant variance explained estimates (7.11–11.76%) reflecting model fit within this cohort. A stepwise regression model highlighted nine candidate SNPs that together explained 51.1% of ALB variance in the study sample. Functional annotation suggested that several variants show eQTL or sQTL signals in tissues relevant to ALB biology (e.g., liver and kidney), and pathway enrichment analyses implicated amino acid and hormone metabolism. Overall, these findings are hypothesis-generating; independent replication in additional and ancestry-matched cohorts (and follow-up functional studies) is required to confirm the robustness of the associations and clarify causal mechanisms. Full article

Short-form video platforms such as TikTok (TikTok Pte. Ltd., Singapore) host large volumes of user-generated, often ephemeral, content related to irregular migration, where relevant cues are distributed across visual scenes, on-screen text, and multilingual captions. Automatically identifying migration-related videos is challenging due to this multimodal complexity and the scarcity of labeled data in sensitive domains. This paper presents an interpretable multimodal classification framework designed for deployment under data-scarce conditions. We extract features from platform metadata, automated video analysis (Google Cloud Video Intelligence), and Optical Character Recognition (OCR) text, and compare text-only, OCR-only, and vision-only baselines against a multimodal fusion approach using Logistic Regression, Random Forest, and XGBoost. In this pilot study, multimodal fusion consistently improves class separation over single-modality models, achieving an F1-score of 0.92 for the migration-related class under stratified cross-validation. Given the limited sample size, these results are interpreted as evidence of feature separability rather than definitive generalization. Feature importance and SHAP analyses identify OCR-derived keywords, maritime cues, and regional indicators as the most influential predictors. To assess robustness under data scarcity, we apply SMOTE to synthetically expand the training set to 500 samples and evaluate performance on a small held-out set of real videos, observing stable results that further support feature-level robustness. Finally, we demonstrate scalability by constructing a weakly labeled corpus of 600 videos using the identified multimodal cues, highlighting the suitability of the proposed feature set for weakly supervised monitoring at scale. Overall, this work serves as a methodological blueprint for building interpretable multimodal monitoring pipelines in sensitive, low-resource settings. Full article

Objective: Since specific bioelectrical reference values for Italian adults are lacking, this study aims to define specific values and test their suitability in pathological cases and athletes. Methods: A sample of 1049 Italian individuals (441 men, 608 women) aged 30–65 years was considered. Competitive athletes (bodybuilding, streetlifting and tennis) were identified within the general sample, and an independent group of individuals with obesity or anorexia nervosa was analyzed for comparison. Anthropometric (weight, kg; stature, mid-upper arm, waist and calf circumferences; cm) and bioelectrical (resistance and reactance, at 50 kHz) variables were taken. Resistivity, (Rsp, Ωcm), reactivity (Xcsp, Ωcm), impedivity (Zsp, Ωcm) and phase angle (PhA, °) were calculated. Two-way ANOVA and Hotelling’s T² test were applied to assess group differences. These data were then pooled with existing datasets to create a comprehensive reference for individuals aged 18 to 100 years. Results: The specific bioelectrical variables were: Rsp = 352.3 ± 55.5, Xcsp = 41.8 ± 9.1, PhA = 6.8 ± 1.0, r (Rsp, Xcsp) = 0.67 (men); Rsp = 384.9 ± 71.2, Xcsp = 40.7 ± 9.4, PhA = 6.1 ± 1.0, r (Rsp, Xcsp) = 0.72 (women). Men showed higher PhA values (p < 0.001), reflecting higher muscle mass and quality, and shorter vectors (p < 0.001), indicative of lower relative fat mass (FM%), than women. Advancing age was associated with lower PhA and longer vectors (p < 0.001). Bioelectrical vectors of individuals with obesity or anorexia nervosa were outside the 95% variability, indicating abnormal values of FM%, whereas those of athletes fell within the lower left quadrant. Conclusions: The specific tolerance ellipses for the Italian adult population fill a gap in the existing literature, providing essential new tools for evaluating body composition in clinical and sports settings, and for comparative analyses. Full article

We used synchrotron X-ray diffraction (XRD) and ac-impedance spectroscopy (AIS) to uncover the structural and chemical modifications undergone by RbH₂PO₄ (RDP) at intermediate temperatures (150 °C < T < 300 °C) and investigate their relationship with RDP’s proton conductivity, σ. Nyquist plots collected on RDP samples sealed in a small volume (~50 mL) of dry air show a gradual increase in σ upon heating from 180 to 260 °C, but not the three-order-of-magnitude superprotonic jump observed in the Cs-based compound CsH₂PO₄ (CDP) within the same temperature range. Correspondingly, XRD measurements using synchrotron radiation (λ = 0.922 Å) on RDP crystalline powders sealed in a quartz capillary exhibit no evidence of a monoclinic-to-cubic superprotonic phase transition like the one observed in CDP. Instead, these temperature-resolved powder XRD patterns demonstrate that the intermediate-temperature RDP monoclinic phase (P2₁/m, a = 7.733 Å, b = 6.189 Å, c = 4.793 Å, and β = 109.21 deg) persists up to the melting point of the title compound. Our most significant finding comes from heating RDP under high pressure (P = 1 GPa), which leads to markedly different structural behavior. Indeed, our full profile refinements against XRD data collected on RDP crystals compressed at ~1 GPa show evidence of a polymorphic phase transition (at T_c = 300 °C) to a high-temperature cubic phase (Pm-3m, a = 4.784 Å) that is isomorphic with its CDP counterpart. This is significant, as it indicates that the superprotonic conduction in phosphate solid acids is not cation-specific, and a general highly efficient proton conduction mechanism is present in the high-temperature phases of these materials. Full article

Exosomes, acting as vital mediators of cellular communication and carriers of diverse biomolecular cargo, are increasingly documented as important participants in cancer pathogenesis and progression. When it comes to triple-negative breast cancer (TNBC), a disease that comes with significant therapeutic hurdles, finding new, non-invasive biomarkers is absolutely crucial. This systematic review considers recent research, focusing on the role of exosomal biomarkers in diagnosing, predicting prognosis and foreseeing treatment response in TNBC patients. After an extensive search across PubMed and Google Scholar, we found many exosomal molecules showing great promise for early detection, tracking disease progression and tailoring treatments. This truly highlights liquid biopsy as a valuable, minimally invasive tool. However, there are still some big challenges to treat. These include variations in methodology, the sheer diversity of samples studied and the prevalence of research in specific populations, all of which make it harder to generalize findings. It has been suggested that future research must prioritize protocol standardization, achieving a deeper understanding of underlying biological mechanisms and, crucially, developing combinatorial biomarker panels. Ultimately, the successful translation of exosomal biomarkers into clinical practice will significantly advance personalized medicine in TNBC, leading to improved patient outcomes and an enhanced quality of life. Full article

Background: Cancer-related health outcomes are shaped by the interplay of aging, complex treatment exposures, and individual psychological characteristics. Mitochondrial dysfunction has been implicated as an underlying biological process affecting cancer-related outcomes. This secondary, exploratory pilot analysis aimed to examine age- and treatment-related differences in fatigue, coping self-efficacy, resilience, and skeletal muscle mitochondrial oxidative capacity, measured via phosphorus-31 magnetic resonance spectroscopy (³¹P-MRS). Methods: Eleven cancer survivors (mean age 53.3 ± 12.7 years) were recruited from a larger symptom management trial. Participants underwent ³¹P-MRS to assess mitochondrial function via phosphocreatine recovery time constant (τPCr). Patient-reported outcome measures and physical function assessments were collected. Group comparisons and correlation analyses were conducted to evaluate differences and associations based on age (<65 vs. ≥65 years) and treatment. Because treatment categories were not mutually exclusive and the time since last treatment was not collected, treatment-related comparisons are descriptive only. Given the small available sample size, we conducted this study as exploratory and hypothesis-generating. Results: Older survivors (≥65) had longer τPCr (59.5 vs. 50.1 s), weaker grip strength, higher fatigue, and lower physical performance compared to younger participants, although differences were not statistically significant. Treatment-related patterns were descriptive; participants receiving multiple treatments had shorter τPCr but lower muscular strength, while immunotherapy recipients reported higher fatigue and lower physical activity. Among younger participants, a negative correlation was observed between τPCr and fatigue (ρ = −0.71), and positive correlations were observed with resilience (ρ = 0.61) and coping self-efficacy (ρ = 0.74), reflecting a pattern that warrants cautious interpretation in this small sample. Conclusions: These preliminary results suggest age- and treatment-related differences in fatigue, physical performance, psychological factors, and skeletal muscle mitochondrial bioenergetics. These signals warrant further testing in larger, adequately powered cohorts to clarify mechanisms and inform the development of personalized survivorship care strategies. Full article

Secure compressive sensing (SCS) mostly benefits scenes such as IoT with finite computer resources, the fields of spaceflight and medicine, etc. Recently, research on SCS has aroused widespread interest. Nevertheless, existing work on embedding security of CS usually requires an extra cryptographic routine applied to the measurement vectors. In this paper, we proposed an SCS scheme boosted by the hyper-chaotic system, which outperforms state-of-the-art methods and endows the SCS with a high level of inherent security. Encryption and sampling processing are accomplished simultaneously in our scheme, i.e., security is achieved when sampling with a measurement matrix, which is generated by an initial-value (secret key)-driven discrete hyper-chaotic (HC) system. Moreover, the application of the HC matrix decreases both the computing and bandwidth consumption costs of secret key streams transmission compared with traditional CS-based encryption methods. Experimentally, the HC-based matrix demonstrates excellent reconstruction performance, achieving an average SSIM of 0.91 and PSNR of 29.09 dB on the Set5 dataset at a sampling ratio of 0.5, outperforming conventional matrices such as Bernoulli and Hadamard. Security analysis confirms that the system exhibits asymptotic spherical secrecy and high key sensitivity—a deviation of ${10}^{-16}$ in the initial value results in complete decryption failure. Furthermore, the scheme shows strong robustness against additive Gaussian white noise and cropping attacks, maintaining a PSNR above 15 dB even under 50% cropping. Compared to existing methods, the proposed approach reduces bandwidth consumption by transmitting only the HC initial parameters rather than the entire measurement matrix. These results demonstrate that the HC-driven SCS framework provides inherent security, high reconstruction fidelity, and practical efficiency, making it suitable for secure sensing in constrained environments. Full article

As an emerging learning paradigm, self-supervised learning (SSL) has attracted extensive attention due to its ability to mine features with effective representation from massive unlabeled data. In particular, SSL, driven by contrastive learning and masked modeling, shows great potential in general visual tasks. However, because of the diversity of ground target types, the complexity of spectral radiation characteristics, and changes in environmental conditions, existing SSL frameworks exhibit limited feature extraction accuracy and generalization ability when applied to complex remote sensing scenarios. To address this issue, we propose a hybrid SSL framework that integrates the advantages of contrastive learning and masked modeling to extract more robust and reliable features from remote sensing images. The proposed framework includes two parallel branches: one branch uses a contrastive learning strategy to strengthen global feature representation and capture image structural information by constructing positive and negative sample pairs; the other branch adopts a masked modeling strategy, focusing on the fine analysis of local details and predicting the features of masked areas, thereby establishing connections between global and local features. Additionally, to better integrate local and global features, we adopt a hybrid CNN+Transformer architecture, which is particularly suitable for intensive downstream tasks such as semantic segmentation. Extensive experimental results demonstrate that the proposed framework not only exhibits superior feature extraction ability and higher accuracy in small-sample scenarios but also outperforms state-of-the-art mainstream SSL frameworks on large-scale datasets. Full article

Road target detection is a crucial aspect of current research in automotive advanced driver assistance systems and intelligent transportation systems, where accuracy, speed, and lightweight design are key considerations. Compared to various sensors employed in driving assistance systems, millimeter-wave radar offers advantages such as all-weather operation, low hardware cost, strong penetration capability, and the ability to extract rich spatial information about targets. This paper tackles the challenges posed by the characteristics of Range-Angle map data from 77 GHz Frequency-Modulated Continuous Wave radar—namely, non-visible light imagery, abstract representation, rich fine details, and overlapping features. To this end, this paper proposes MMPFNet, a lightweight model based on the hypergraph mechanism with attention enhancement, as an extension of YOLOv13. First, an M-DSC3k2 module is proposed based on the hypergraph mechanism to enhance attention toward small targets. Second, a detection head with a double-bottleneck inverted MBConv-block structure is designed to improve the model’s accuracy and generalization capability. Third, a lightweight PPLConv module is customized to transform the backbone network, enhancing the model’s lightweight design while slightly reducing its accuracy. Considering the differences from traditional visible light datasets, the Focus Expansion-IoU loss function is introduced into the model to focus attention on different regression samples. The MMPFNet model achieves significant improvements in detecting common road targets such as pedestrians, bicycles, cars, and trucks on the Frequency-Modulated Continuous Wave radar Range-Angle dataset compared to the baseline YOLOv13n model: mAP50-95 increases by 16%, precision improves by 6%, and recall rises by 8.7%. MMPFNet is also evaluated on other non-visible light datasets such as CRUW-ONRD and soundprint datasets. Compared to commonly used detection models like FCOS and RetinaNet, MMPFNet achieves significant performance gains, attaining state-of-the-art results. Full article

Reliability is a critical indicator for evaluating the safety and performance of mechanical structures. Surrogate models are widely employed in reliability analysis to reduce computational costs; however, their accuracy strongly depends on the quality of training samples. This paper proposes an uncertainty and distance-driven sequential sampling deep neural network method (UDDS-DNN) for structural reliability analysis. First, initial training samples were generated from the Monte Carlo population (MCP) using the Minimax sampling strategy. This ensured global coverage of the random variable space. Then, the predictive uncertainty of the DNN was quantified using the Jackknife-based method. The estimated uncertainty, inter-sample distance, and probabilistic characteristics of the input random variables were considered. The adaptive learning function was constructed to guide sequential sample enrichment. The proposed method was validated through two numerical examples and three engineering case studies. Results demonstrated that UDDS-DNN achieves high accuracy and computational efficiency in multivariate, nonlinear, and complex structural reliability problems. The method provided an effective solution for reliability analysis of complex structures using DNN-based sequential sampling. Full article

In order to solve the problems of insufficient medical image feature extraction, high classification accuracy, and computational complexity in automatic diagnosis of skin lesions in the edge computing environment, this paper proposes a real-time pseudo-multimodal low-delay diagnosis framework, SCGViT, based on a vision transformer. The framework is constructed around three functional objectives: mitigating data imbalance through generative modeling, capturing diverse representations via multi-dimensional perception, and optimizing feature fusion through adaptive refinement. Firstly, using Class-Conditioned Generative Adversarial Networks (CGANs) simulates manifolds of minority class samples in latent space, achieving preliminary balance of data distribution. Secondly, a branch feature-extraction path is constructed to simulate inversion (INV) and infrared (IR) modes in the original visual primary color mode (RGB), in order to achieve multi-dimensional perception. Finally, a cross-attention mechanism is combined for cross-branch feature aggregation, and a channel-attention mechanism (squeeze and excitation) is embedded for secondary refinement of the mixed global local features to enhance the representation ability of key pathological regions by integrating complementary structural and contrast information. The experimental results on the HAM10000 dataset showed that the F1 score reached 0.973, the inference speed reached 304.439 FPS, the parameter count was only 0.524 M, and the computational complexity was only 0.866 G FLOPs, achieving a balance between high accuracy and light weight. Full article

Background: Navigational bronchoscopy (NB) enables precise sampling of peripheral and central pulmonary nodules using shape-sensing or electromagnetic guidance. A major challenge is anesthesia-induced atelectasis, which alters lung anatomy, reduces registration accuracy, and is known to lower diagnostic accuracy. To counteract this, ventilatory protocols such as the Ventilatory Strategy to Prevent Atelectasis (VESPA) and the Lung Navigation Ventilation Protocol (LNVP) have been recommended. Their adoption and clinical impact, however, remain uncertain. Methods: We conducted a retrospective review of 224 consecutive NB procedures performed under general anesthesia at a single academic medical center (January 2020–August 2024). Demographic, anesthetic, and ventilatory data were extracted from electronic records. Outcomes included navigational success (ability to reach the lesion) and diagnostic accuracy (concordance between bronchoscopic diagnosis and final clinical diagnosis after follow-up). Ventilatory practices were compared with published VESPA and LNVP recommendations. Results: Navigational success, defined as successful advancement of the bronchoscope to the target lesion with tissue acquisition, was achieved in 89.2% of cases. Overall diagnostic accuracy, defined as concordance between bronchoscopic diagnosis and final clinical diagnosis after follow-up, was 81.7%. Ventilatory management consistently diverged from recommended protocols. Most patients were ventilated with FiO₂ > 0.6, PEEP in the range of 7–10 cm H₂O, and tidal volumes of 300–500 mL. The only recommended maneuver systematically applied was recruitment immediately after intubation. Despite widespread deviation from both VESPA and LNVP, diagnostic performance remained favorable relative to published benchmarks. No major anesthesia-related complications occurred. Conclusions: In this retrospective series, navigational success comparable to published studies that adapted strict ventilation protocols was achieved with also comparable diagnostic accuracy without strict adherence to predefined ventilatory strategies. Recruitment maneuvers may represent the most influential component of current protocols, but institutional factors such as procedural expertise and case volume likely contributed to outcomes. Prospective studies are warranted to determine whether standardized ventilatory protocols are necessary for optimizing NB performance. Full article

Rapeseed stalks (Brassica napus), an abundant agricultural residue, represent a promising non-woody raw material for the pulp and paper industry. This study focuses on the chemical and structural characterization of alkali lignins isolated from black liquors generated by two common delignification methods: Kraft and Soda-Anthraquinone Pulping of rapeseed stalks. The objective is to understand how the chemical environment of each process influences the final structure, fragmentation degree, and reactivity of the isolated lignin. In practice, lignin samples are recovered from black liquors produced under varying conditions (alkali charge, time, and temperature) to achieve defined levels of delignification. Detailed characterization was performed using advanced analytical techniques, including Gel Permeation Chromatography, Solid-State Cross-Polarization/Magic-Angle-Spinning Nuclear Magnetic Resonance, and FT-IR and UV-Vis Spectroscopy. The findings provide essential data on the structural differences, confirming the suitability of the resulting materials for potential high-value applications. Furthermore, the structural similarities and performance indicators suggest that the Soda-AQ process enables successful delignification of rapeseed stalks without the sulfur emission issues associated with the Kraft method, thus validating the former as an environmentally cleaner alternative for non-wood biomass utilization supporting the complete valorization of rapeseed agricultural waste. Full article