Search Results (11,248)

Accurate correspondence of complex neuroanatomical surfaces under non-rigid deformations remains a formidable challenge in computational neuroimaging, owing to inter-subject topological variability, partial occlusions, and non-isometric distortions. Here, we introduce the Dynamic Graph Analyzer (DGA), a unified hybrid framework that integrates simplified structural descriptors with spatial constraints and formulates matching as a global linear assignment. Structurally, the DGA computes node-level metrics, degree weighted by betweenness centrality and local clustering coefficients, to capture essential topological patterns at a low computational cost. Spatially, it employs a two-stage scheme that combines global maximum distances and local rescaling of adjacent node separations to preserve geometric fidelity. By embedding these complementary measures into a single cost matrix solved via the Kuhn–Munkres algorithm followed by a refinement of weak correspondences, the DGA ensures a globally optimal correspondence. In benchmark evaluations on the FAUST dataset, the DGA achieved a significant reduction in the mean geodetic reconstruction error compared to spectral graph convolutional netwworks (GCNs)—which learn optimized spectral descriptors akin to classical approaches like heat/wave kernel signatures (HKS/WKS)—and traditional spectral methods. Additional experiments demonstrate robust performance on partial matches in TOSCA and cross-species alignments in SHREC-20, validating resilience to morphological variation and symmetry ambiguities. These results establish the DGA as a scalable and accurate approach for brain shape correspondence, with promising applications in biomarker mapping, developmental studies, and clinical morphometry. Full article

In this paper, we propose a novel visual tracking method with a object part-aware attention-based matching (OPAM) mechanism, which leverages local–global attention to enhance visual tracking performance. Our method introduces three key components: (1) a local part-aware global self-attention mechanism that embeds rich contextual information among candidate regions, enabling the model to capture mutual dependencies and relationships effectively, (2) a local part-aware global cross-attention mechanism that injects target-specific information into candidate region features, improving the alignment and discrimination between the target and background, and (3) a global cross-attention mechanism that extracts object holistic information from the target-search feature context for further discriminability. By integrating these attention modules, our approach achieves robust feature aggregation and precise target localization. Extensive experiments on a large-scale tracking benchmark demonstrate that our method shows competitive performance metrics in both accuracy and robustness, particularly under challenging scenarios such as occlusion and appearance changes, while running at real-time speeds. Full article

This study proposes a couple of analytical solutions that characterize the anisotropic dense celestial bodies within the Rastall-Rainbow theoretical framework. The analysis assumes a static spherically symmetric matter distribution and derives the corresponding modified field equations. By utilizing well-established radial metric functions and merging them with the two principal pressures, we obtain differential equations related to the time component. Subsequently, we perform the integration of these equations to determine the remaining geometric quantity that encompasses various integration constants. The proposed interior solutions are then matched with the Schwarzschild exterior metric at the boundary of the compact object, facilitating the determination of the constants. Additionally, the incorporation of the non-minimal coupling parameter into these constants is accomplished by enforcing the null radial pressure at the boundary. Afterwards, we rigorously examine the physical characteristics and critical stability conditions of the formulated models under observational data from two pulsars, say 4U 1820-30 and LMC X-4. It is concluded that our models are well-aligned with essential criteria required to ensure the physical viability of stellar structures, subject to specific parametric values. Full article

Yin Yang 1 (YY1) is a multifunctional transcription factor implicated in gene regulation, cell proliferation, and survival. While its role in breast cancer (BC) has been explored, its prognostic significance remains controversial. In this study, we evaluated nuclear YY1 expression in 276 BC tissue samples using immunohistochemistry (IHC), tissue microarrays (TMAs), and digital pathology (DP). Nuclear staining was quantified using Aperio ImageScope software, focusing on tumor regions to avoid confounding from stromal or non-tumor tissues. This selective and standardized approach enabled precise quantification of YY1 expression. Our results show elevated median YY1 expression in tumor vs. normal matched tissues (p < 0.001). The optimal cutoff for medium-intensity nuclear YY1 expression in tumor areas for overall survival (OS) was established by a receiver operating characteristic (ROC) curve (AUC = 0.718, 95% CI: 0.587–0.849, p = 0.008). In contrast, ROC curves showed no prognostic impact (AUC and p-value) for YY1 quantification in whole spots (tumor + normal). As a categorical variable, high YY1 expression was correlated with more aggressive BC features, including tumor size > 3 cm (57.7% vs. 44.2% p = 0.037), the triple-negative breast cancer (TNBC) molecular subtype (27.3% vs. 13.9% p = 0.026), and advanced prognostic stage (III) (31.8% vs. 16.7% p = 0.003), while as a continuous variable, YY1 was associated with higher histological (p = 0.003) and nuclear grades (p = 0.022). High YY1 expression was significantly associated with a reduced OS of BC patients, as shown by Kaplan–Meier curves (HR = 2.227, p = 0.002). Since YY1 was significantly enriched in TNBC, we evaluated its prognostic resolution in this subgroup. But, probably due to the small number of patients within this subset, our results were not statistically significant (HR = 1.317, 95% CI: 0.510–3.405, p = 0.566). Next, we performed multivariate Cox regression, confirming YY1 as an independent prognostic factor for overall survival (HR = 1.927, 95% CI: 1.144–3.247, p = 0.014). In order to improve prognostic value, we constructed a mathematical model derived from the multivariate Cox regression results, including YYI, AJCC prognostic stage (STA), and axillary lymph node dissection (ALN), with the following equation: h(t) = h₀(t) × exp (0.695 × YY1 + 1.103 × STA − 0.503 × ALN). ROC analysis of this model showed a better AUC of 0.915, similar sensitivity (83.3%), and much higher specificity (92%). Bioinformatic analysis of public datasets supported these findings in BC, showing YY1 overexpression in multiple cancer types and its association with poor outcomes in BC. These results suggest that YY1 may play a role in tumor progression and serve as a valuable prognostic biomarker in BC. DP combined with molecular data enhanced biomarker accuracy, supporting clinical applications of YY1 in routine diagnostics and personalized therapy. Additionally, developing a combined score based on the modeling of multiple prognostic factors significantly enhanced survival predictions, representing a practical tool for risk stratification and the guidance of therapeutic decisions. Full article

This study assesses how classical and modern features extracted from a single ECG lead (II) influence automated arrhythmia classification. Using the Large Scale 12-Lead Electrocardiogram Database for Arrhythmia Study and MATLAB^®, we compared traditional morphological measures (e.g., QRS duration, QT interval, atrial/ventricular rates) with advanced time-, frequency-, and nonlinear-domain descriptors. The method classifies ECGs into four or eight categories using 15–39 features, either automatically selected or combined. In the eight-class task, 29–39 features yielded 69% accuracy; in the four-class task, 15 MRMR-selected features achieved 94.2% accuracy. A key strength is efficiency: relying on a single lead reduces preprocessing, storage, and classification time by a factor of ~12 compared with 12-lead approaches. These findings show that advanced descriptors from a single lead can match multi-lead performance, enabling practical, scalable clinical applications. Full article

To directly address the important issue of MRI geometric distortions in stereotactic radiosurgery (SRS) planning, we performed a phantom study of sequence acquisition optimization. This study analyzed, in particular, the effects of clinically relevant gadolinium (Gd) concentration as filling solution for the phantom, as well as phase encoding reversal direction and flip angle on distortion. We created a rigid geometric grid phantom with 840 fiducial markers for distortion quantification on a 3T MRI scanner. To choose the optimal filling solution, an anthropomorphic RANDO phantom was employed, and 1 mmol/L gadolinium was chosen due to clinical relevance. An automated Python-based software (version 3.7.1) was developed for efficient detection and matching of phantom inserts between MRI and CT scans. A series of MRI acquisition parameter optimizations were systematically evaluated. The standard SRS protocol exhibited the highest average distortion of 1.301 mm. Notably, reversing the phase-encoding direction to anterior–posterior (AP) reduced the mean distortion to 0.725 mm, a 44.27% decrease, while the maximum distortion was reduced by 15.65%. The AP phase sequence maintained acquisition time, SAR, SNR, and CNR within acceptable limits. Additional distortion reduction was achieved by increasing the flip angle from 12° to 18°. In this work, we succeeded in significantly reducing the mean distortion observed in phantom images. As the gadolinium concentration used in the phantom is clinically similar to the gadolinium concentration observed in patients undergoing MRI scans with contrast agents, the achieved distortion reduction is prospectively reproducible in patients. Full article

Current mainstream remote sensing target detection algorithms mostly estimate the rotation angle of targets by designing different bounding box descriptions and loss functions. However, they fail to consider the symmetry–asymmetry duality anisotropy in the distribution of key features required for target localization. Moreover, the equivalent feature extraction mode of shared convolutional kernels may lead to difficulties in accurately predicting parameters with different attributes, thereby reducing the performance of the detector. In this paper, we propose the Feature Equalization and Hierarchical Decoupling Network (FEHD-Net), which comprises three core components: a Symmetry-Enhanced Parallel Interleaved Convolution Module (PICM), a Parameter Decoupling Module (PDM), and a Critical Feature Matching Loss Function (CFM-Loss). PICM captures diverse spatial features over long distances by integrating square convolution and multi-branch continuous orthogonal large kernel strip convolution sequences, thereby enhancing the network’s capability in processing long-distance spatial information. PDM decomposes feature maps with different properties and assigns them to different regression branches to estimate the parameters of the target’s rotating bounding box. Finally, to stabilize the training of anchors with different qualities that have captured the key features required for detection, CFM-Loss utilizes the intersection ratio between anchors and true value labels, as well as the uncertainty of convolutional regression during training, and designs an alignment criterion (symmetry-aware alignment) to evaluate the regression ability of different anchors. This enables the network to fine-tune the processing of templates with different qualities, achieving stable training of the network. A large number of experiments demonstrate that compared with existing methods, FEHD-Net can achieve state-of-the-art performance on DOTA, HRSC2016, and UCAS-AOD datasets. Full article

Addressing the challenges of insufficient incentives, weak constraints, and superficial adoption in promoting human–robot collaboration (HRC) in the construction industry, this study develops a tripartite evolutionary game model among government, contractors, and on-site teams under bounded rationality. Lyapunov stability analysis and numerical simulation are employed to conduct parameter sensitivity analyses. The results show that a strategy profile characterized by flexible regulation, deep adoption, and high-effort collaboration constitutes a stable evolutionary outcome. Moderately increasing government incentives helps accelerate convergence but exhibits diminishing returns under fiscal constraints, indicating that subsidies alone cannot sustain genuine engagement. Reducing penalties for contractors and on-site teams, respectively, induces superficial adoption and low effort, whereas strengthening penalties for bilateral violations simultaneously compresses the space for opportunistic behavior. When the payoff advantage of deep adoption narrows or the payoff from perfunctory adoption rises, convergence toward the preferred steady state slows markedly. Based on the discussion and simulation evidence, we recommend dynamically matching incentives, sanctions, and performance feedback: prioritizing flexible regulation to reduce institutional frictions, configuring differentiated sanctions to maintain a positive payoff differential, reinforcing observable performance to stabilize frontline effort, and adjusting policy weights by project stage and actor characteristics. The study delineates how parameter changes propagate through behavioral choices to shape collaborative performance, providing actionable guidance for policy design and project governance in advancing HRC. Full article

Traditional security methods fail to match the speed of evolving threats because Industrial Internet of Things (IIoT) technologies have become more widely adopted. A lightweight adaptive AI-based intrusion detection system (IDS) for IIoT environments is presented in this paper. The proposed system detects cyber threats in real time through an ensemble of online learning models that also adapt to changing network behavior. The system implements SHAP (SHapley Additive exPlanations) for model prediction explanations to allow human operators to verify and understand alert causes while addressing the essential need for trust and transparency. The system validation was performed using the ToN_IoT and Bot-IoT benchmark datasets. The proposed system detects threats with 96.4% accuracy while producing 2.1% false positives and requiring 35 ms on average for detection on edge devices with limited resources. Security analysts can understand model decisions through SHAP analysis because packet size and protocol type and device activity patterns strongly affect model predictions. The system underwent testing on a Raspberry Pi 5-based IIoT testbed to evaluate its deployability in real-world scenarios through emulation of practical edge environments with constrained computational resources. The research unites real-time adaptability with explainability and low-latency performance in an IDS framework specifically designed for industrial IoT security. The solution provides a scalable method to boost cyber resilience in manufacturing, together with energy and critical infrastructure sectors. By enabling fast, interpretable, and low-latency intrusion detection directly on edge devices, this solution enhances cyber resilience in critical sectors such as manufacturing, energy, and infrastructure, where timely and trustworthy threat responses are essential to maintaining operational continuity and safety. Full article

Background: Missed miscarriage (MM) is a common first-trimester complication. Misoprostol alone achieves moderate success, while combination therapy with mifepristone improves outcomes in spontaneous pregnancies. Evidence in assisted reproductive technology (ART) pregnancies is scarce. We evaluated whether combined mifepristone–misoprostol improves outcomes in ART pregnancies compared with misoprostol alone and compared results with spontaneously conceived (SC) pregnancies. Methods: This retrospective matched cohort study was conducted at a single center (2017–2024). ART pregnancies were matched 1:2 with SC pregnancies by maternal age. Patients received misoprostol alone or 200 mg mifepristone followed 48 h later by misoprostol. The primary outcome was treatment success, defined as complete uterine evacuation without repeat misoprostol or surgery. Secondary outcomes included emergency visits, surgical procedures, and ART-related predictors. Subgroup analyses were performed by ART protocol. Results: Among 307 patients (94 ART, 213 SC), combined therapy yielded higher success than misoprostol alone in SC (84% vs. 71%, p = 0.023) and ART pregnancies (95% vs. 80%, p = 0.035). In hormonally supported frozen embryo transfer (HRT-FET) cycles, success was 100% with combined therapy versus 80% with misoprostol alone. Conclusions: Combined mifepristone–misoprostol is more effective than misoprostol alone, with particularly high success in HRT-FET cycles. Full article

Objetives: This study aimed to (1) describe and compare the external physical requirements of international cerebral palsy (CP) football players during training sessions and official matches at the 2024 IFCPF World Cup, and (2) analyze the relationships between standardized field-based physical performance tests and the physical requirements recorded in both contexts. Methods: Twelve international outfield players from the Spanish national CP football team were monitored throughout the tournament. Physical performance was evaluated two weeks prior using 5-m and 30-m sprints, a Modified Agility Test (MAT), a dribbling test, and the 30–15 Intermittent Fitness Test (vIFT). Match and training physical requirements were assessed using inertial devices, including total and relative distances, velocity metrics, and acceleration/deceleration outputs. Results: Matches imposed significantly greater demands than training sessions in terms of peak velocity, total distance per minute, and distance at moderate (>12–18 km/h) and high (>18 km/h) intensities (t = 2.79 to 8.06; p = 0.01; ES(d) = 0.50 to 1.45). Training sessions exhibited greater variability in load while match requirements were consistent across games. Performance in the MAT and dribbling tests correlated with several physical indicators in both training and competition. In contrast, vIFT and sprint tests showed limited associations, especially with match variables. Conclusions: Match play elicits higher and more stable physical requirements than training. The MAT and dribbling tests appear to be ecologically valid tools for assessing functional readiness in CP football. These findings support the integration of specific physical tests and tailored training designs to better replicate the competitive requirements of international CP football. Full article

Circular agriculture reclaims nutrients from waste streams to reduce fertilizer imports, mitigate environmental impacts, and close material loops. This review evaluates the agronomic performance of nitrogen, phosphorus, and potassium products recovered from wastewater, crop residues, and manure compared with conventional fertilizers. A structured literature survey identified 85 pot and field trials published between 2010 and 2024, covering ammonium salts, struvite, ashes, compost, digestate, biochar, hydrochar, and biostimulants. Ammonium sulfate and nitrate consistently matched synthetic yields (95–105%) due to their solubility and immediate N availability, while aqueous ammonia showed variable results depending on application timing and soil pH. Struvite and phosphorus-rich ashes performed best (90–100%) in neutral to slightly acidic soils, whereas organo-mineral phosphate fertilizers (85–95%) were less effective in alkaline soils. Potassium-rich ashes and waste mica were effective (80–95%) in soils with moderate cation exchange, though mica underperformed (60–75%) in coarse soils. Biochars and hydrochars improved soil water retention and nutrient exchange, yielding 90–110% of synthetic performance, while biostimulants increased yields by 8–20%. Recovered products demonstrate agronomic equivalence while offering co-benefits for soil health, waste management, and circular economy goals. Future work should prioritize long-term field validation, techno-economic analysis, and regulatory integration to enable large-scale adoption. Full article

This study presents a new method for estimating the seismic responses of multi-story structures equipped with a cost-effective earthquake protection system. This system comprises a graphite lubrication interface, targeting a friction coefficient of approximately 0.2, and a feasible restoring force mechanism to suppress residual displacements. It utilizes the concept of sliding systems through conventional and affordable construction materials although it acts like a fixed-based structure until exceeding the threshold level. This multi-story estimation procedure is an extension of the recently developed procedure for estimating the shear coefficient of a single-story sliding structure with a restoring force mechanism. In the new estimation procedure, a multi-story superstructure is firstly regarded as a single-story superstructure to determine the shear coefficient. Then, the shear coefficient is distributed to each story through floor distribution coefficients considering the mass ratios. The contribution of ground motion intensity is also incorporated into the new form for improving accuracy. For this examination, incremental dynamic analyses (IDAs) are performed for three and six-story free-standing structures, both with and without a restoring force capability. The results clarify the reliability of the new estimation, which matched the IDA results within the ±20% error. The improvement in accuracy achieved by incorporating ground motion intensity is also clarified. The multi-story estimation with the improvement can reasonably estimate the seismic response of sliding structures, without dynamic analysis, solely based on structural properties. This greatly benefits the design process. Furthermore, the IDA results clarified the significant benefits of multi-story sliding structures employing graphite lubrication and properly designed restoring force mechanisms in reducing structural damage and suppressing residual sliding displacements. Full article

Physical fitness, especially strength, is key for football performance and injury prevention, but its role in match running remains unclear. This study examined correlations between knee flexor and extensor isokinetic torque (60°/s, 180°/s, 300°/s) and running performance in 20 U17 players (age 16 ± 0.5 years, height 1.78 ± 0.05 m, weight 71 ± 7.1 kg) across 16 matches, with GPS tracking (Apex, STATSports, Newry, Northern Ireland). Results varied by position. Central defenders showed a negative correlation between non-dominant knee flexor torque at 300°/s and high-speed running distance (r = −0.975, p = 0.025). Side defenders displayed positive correlations between dominant knee flexor torque at 300°/s and both total distance and moderate running (r = 0.885–0.976, p < 0.05), but negative correlations with maximum speed (r < −0.89, p < 0.05). Central midfielders had several negative associations between dominant knee flexor torque at 300°/s and accelerations or decelerations (r < −0.88, p < 0.05). Side midfielders and forwards showed positive correlations between torque at higher speeds (180°/s, 300°/s) and sprint distance (r ≥ 0.85, p < 0.05). Overall, associations differed by position, velocity, and performance variable, reflecting tactical demands and the limits of single-joint testing. However, in SMFs, high positive correlations were observed, highlighting the importance of strength and its necessity in the training process. Full article

Few-shot instance segmentation (FSIS) is devised to address the intricate challenge of instance segmentation when labeled data for novel classes is scant. Nevertheless, existing methodologies encounter notable constraints in the agile expansion of novel classes and the management of memory overhead. The integration workflow for novel classes is inflexible, and given the necessity of retaining class exemplars during both training and inference stages, considerable memory consumption ensues. To surmount these challenges, this study introduces an innovative framework encompassing a two-stage “base training-novel class fine-tuning” paradigm. It acquires discriminative instance-level embedding representations. Concretely, instance embeddings are aggregated into class prototypes, and the storage of embedding vectors as opposed to images inherently mitigates the issue of memory overload. Via a Region of Interest (RoI)-level cosine similarity matching mechanism, the flexible augmentation of novel classes is realized, devoid of the requirement for supplementary training and independent of historical data. Experimental validations attest that this approach significantly outperforms state-of-the-art techniques in mainstream benchmark evaluations. More crucially, its memory-optimized attributes facilitate, for the first time, the conjoint assessment of FSIS performance across all classes within the COCO dataset. Visualized instances (incorporating colored masks and class annotations of objects across diverse scenarios) further substantiate the efficacy of the method in real-world complex contexts. Full article