Search Results (2,016)

Surface-enhanced Raman scattering (SERS) offers a powerful route for biomolecular detection because it combines molecular specificity with high sensitivity, rapid optical readout, and multiplexing capability. In real biological samples, however, analytical performance is rarely determined by signal enhancement alone. Biofluids such as serum, plasma, saliva, urine, and interstitial fluid contain complex biomolecular mixtures that interfere with target capture, spectral response, and data interpretation. A practical SERS biosensor must therefore localize targets, stabilize spectral responses, tolerate matrix-induced variation, and convert complex spectra into reliable analytical information. This review discusses recent progress in SERS biosensing from an integrated system perspective, with particular focus on artificial intelligence/machine learning (AI/ML)-assisted interpretation. Direct label-free SERS provides chemically transparent readouts but is limited by stochastic adsorption, hotspot heterogeneity, and spectral variation in complex samples. Bio-recognition interfaces improve target localization, while signal-transduction strategies based on nanotags, immunoassays, clustered regularly interspaced short palindromic repeats (CRISPR) systems, nanozymes, and lateral-flow formats decouple molecular recognition from spectral generation. Digital SERS further improves measurement robustness by converting fluctuating intensities into countable, event-based outputs. AI/ML-assisted analysis can support full-spectrum classification, calibration transfer, explainability, and patient-level decision-making. We frame AI/ML-assisted SERS biosensing as an integrated architecture connecting substrate design, interface engineering, signal transduction, digital measurement, and clinical validation. Future progress will depend as much on validation-ready workflows as on plasmonic enhancement itself, especially for systems intended to operate across different samples, instruments, and clinical settings. Full article

Background/Objectives: Hysteroscopy is the gold standard for visualization and treatment of intrauterine pathology. Because hysteroscopic interpretation remains operator-dependent, artificial intelligence (AI) has been evaluated as a tool to improve consistency, lesion recognition, and decision support. We aimed to systematically review AI, machine learning (ML), deep learning (DL), or computer-aided diagnosis (CAD) applications in hysteroscopy. Methods: A systematic search of PubMed/MEDLINE and EBSCOhost was performed from database inception to 8 March 2026, supplemented by targeted searches. Risk of bias was assessed using QUADAS-2 (diagnostic), PROBAST (prognostic), RoB2, and structured technical quality domains. Results: Nineteen primary studies were included, covering five areas: diagnostic classification and object detection (n = 8), real-time lesion detection and localization (n = 4), segmentation and visual-field support (n = 3), operative guidance (n = 1), and prognostic or decision-support applications (n = 3). Performance was highest in narrowly defined binary tasks and in large multicenter systems (e.g., ECCADx: AUC 0.979 internal, 0.975 external) and in prognostic fertility-prediction models after hysteroscopic adhesiolysis (AUC up to 0.992). Broader multiclass classification of heterogeneous lesions showed uneven and lower performance. Most studies were single-center, retrospective, and lacked external validation. Only one randomized study linked AI support to measurable procedural outcomes. Conclusions: The available studies indicate good technical performance in selected hysteroscopic tasks, particularly binary classification, focal lesion detection, and postoperative fertility stratification. Current evidence, however, remains limited by retrospective design, operator-dependent image acquisition, inconsistent validation, and scarce outcome-based clinical testing. In the short term, the most likely role of these systems is to support image interpretation, improve visual quality control, highlight suspicious lesions, and integrate hysteroscopic findings with complementary clinical data. Full article

Dynamic hospital corridor environments are characterized by complex corridor environments, diverse target-scale variations, frequent occlusions, and dense small-object distribution, posing significant challenges to the accuracy and efficiency of the existing methods on resource-constrained platforms. To effectively address these challenges, a high-precision framework CSAWOA (Cross Search Adaptive Whale Optimization Algorithm)-YOLOv8 (You Only Look Once version 8) model for complex medical environments was introduced in this work. By jointly modelling high-level semantic information and low-level cues such as texture and colour, the proposed model achieved a more discriminative and informative feature representation. The T-CBS (Transformer-Convolutional Bottleneck Structure) module, capable of extracting shallow-level features and integrating global contextual information to address target occlusion issues, was also proposed. Furthermore, the integration of the BiFormer module yielded an enhanced feature discriminability, improving small-target recognition while reducing sensitivity to background noise. The classification function was modified, effectively solving the problem of class imbalance in complex corridor environments. The combination of these two concepts achieved an effective balance of diversity in detection and convergence speed, leading to improved optimization performance and greater resistance to local-optimum stagnation. Meanwhile, an improved version of the WOA was developed, termed CSAWOA, enabling automatic hyperparameter optimization for the improved YOLOv8 model. From the experimental results, improvements of 4.9%, 6.1%, and 8.3% in mAP, precision, and recall, respectively, compared to YOLOv8 were demonstrated, while also exhibiting better generalization. Overall, the proposed method provides a reliable and efficient approach for object detection in complex hospital corridors, offering a valuable foundation for future research and real-world healthcare applications. Full article

The introduction of noise during image acquisition and transmission is inevitable, leading to a significant reduction in the accuracy of image processing tasks, such as target classification, localization, and recognition. To address this issue, this paper proposes a novel robustness-oriented pooling module called Gaussian adaptive pooling. Drawing on the principles of Gaussian filters, the method introduces a Gaussian weight for feature values in the pooling operation, thus integrating filtering and pooling in a novel manner. This approach is both lightweight and versatile, requiring no additional learnable parameters, and enables seamless integration into neural network architectures with pooling layers. Rigorous mathematical derivations and simulation experiments show that our proposed Gaussian adaptive pooling method surpasses conventional methods (average-pooling and max-pooling) in noise handling. Furthermore, its robustness is comparable to traditional pooling methods in addressing challenges such as rotations, scalings, and translations. Extensive evaluations across multiple computer vision tasks—including image classification (CIFAR-10/100), object detection (MS COCO and RTTS), and semantic segmentation (CamVid)—confirm its effectiveness. Specifically, under varying levels of noise and degraded conditions, Gaussian adaptive pooling achieves significant improvements in standard performance metrics compared to conventional pooling methods. For instance, it delivers notable quantitative gains across different tasks including up to a 12.67% increase in mean intersection over union on the CamVid dataset for semantic segmentation and a 1.1% mAP50 enhancement on the real-world RTTS dataset for object detection. Full article

Background/Objectives: Gastric cancer is a major cause of morbidity and mortality in Saudi Arabia, where late-stage presentation is common. Public awareness of risk factors and warning symptoms is essential for early detection. This study assessed gastric cancer knowledge and its demographic correlates among adults in Jeddah. Methods: A cross-sectional survey of 1400 adults was conducted between August and October 2025 using a structured 45-item questionnaire covering 37 knowledge items across four domains. Analyses used independent t-tests, one-way ANOVA, chi-squared tests, and multivariable linear regression. Results: The mean age was 31.34 ± 10.18 years; 53.8% were women. The mean total knowledge score was 18.94 ± 7.34/37 (51.2%). Domain scores were risk factors 4.95 ± 2.04/11 (45.0%), symptoms 4.54 ± 2.87/10 (45.4%), prevention 4.91 ± 2.09/10 (49.1%), and management 2.99 ± 1.95/6 (49.8%). Knowledge was associated with occupation (p < 0.001), with healthcare workers scoring highest (25.97 ± 5.56). Higher knowledge was also associated with education (p < 0.001), age (p = 0.003), and family or friend history of gastric cancer (p = 0.003), but not sex (p = 0.39). In multivariable analysis, educational attainment (β = 0.30) and healthcare-provider occupation (β = 0.23) were the strongest independent correlates of knowledge (both p < 0.001). Only 34% of respondents identified Helicobacter pylori as a risk factor. Conclusions: Awareness of gastric cancer in Jeddah is suboptimal, particularly for risk factors and symptom recognition. Targeted interventions addressing risk factors, alarm symptoms, and dietary risks—delivered through primary care and community channels—may improve awareness and support earlier detection. Full article

Community detection in attributed networks becomes considerably more challenging when the number of communities is unknown in advance. Most existing deep community detection methods assume a fixed community count, whereas reinforcement learning (RL)-based alternatives often suffer from overestimated action values and unstable target updates. To address these limitations, we propose GCR-Net (Graph Community Recognition Network), an RL-guided framework that combines representation learning with adaptive community-count selection. The method adapts decoupled value estimation and gradual anchor-network updates from established deep RL techniques to a formal MDP over candidate community counts. Experiments on citation, social, biomedical, and proteininteraction benchmarks, together with synthetic graphs with more than ten communities, show that GCR-Net delivers competitive NMI and ARI scores with lower variance and more stable optimization than conventional RL baselines. Statistical tests indicate that the clearest gains concern training stability rather than large accuracy margins. Full article

This paper presents the RaDICAL monostatic passive radar framework for target detection and localization using a sparse uniform circular array (SUCA), multifrequency dither, and dictionary-based waveform processing. Rather than forming conventional spatial images or relying on explicit Doppler/TDOA/FDOA estimation, the proposed method encodes target geometry directly into a composite receiver waveform and performs localization through hypothesis testing using a library of predicted waveform responses. A SUCA-based signal model is developed for both point and extended targets, and detection/localization formulated as a waveform-domain dictionary matching problem using normalized complex correlation and QR-domain processing. A reproducible MATLAB-based Monte Carlo study evaluates waveform separability, probability of detection versus input SNR, receiver operating characteristic (ROC) behavior, localization performance, and receiver power balance. The results demonstrate that multifrequency dither produces distinctive composite waveforms with strong hypothesis separability and stable waveform domain recognition performance. ROC analysis and detection simulations showed reliable target detection at input SNR levels on the order of −10 to 0 dB, consistent with the coherent processing gain achieved through waveform-domain correlation processing. The corresponding power-balance analysis indicates that reliable detection and localization are feasible using modest illuminator EIRP and compact receiver dimensions. These results support the feasibility of compact reference-free waveform domain passive sensing for joint target detection and localization. Full article

Aircraft cargo compartment fires constitute a significant type of aviation fire, posing a grave threat to aviation safety. To guard against and respond to such fires, existing aircraft cargo compartments are equipped with smoke detection fire detectors, which rely on perceiving changes in smoke transmittance to determine the onset of a fire. However, these detectors offer relatively low recognition accuracy and cannot provide a direct visual representation of the fire. In this work, we introduce a fire recognition method built on image sensors and a deep learning model. In light of the irregular shapes of flames and smoke, an improved interactive triplet attention mechanism (ITAM) is integrated into the You Only Look Once version 5 (YOLOv5) model, enhancing the model’s recognition accuracy. Furthermore, the original Neck structure is replaced with an Asymptotic Feature Pyramid Network (AFPN), improving the model’s ability to recognize small targets, which is particularly useful for detecting flames and smoke early in a fire. This paper further improves the model’s recognition accuracy by introducing the Focaler-IoU loss function, which balances the feature learning of hard and easy samples. Therefore, the network model in this paper is named IATF-YOLO. Ablation experiments demonstrate that our algorithm improves accuracy by 2%, while comparative experiments with several mainstream baseline models show that our algorithm achieves a 0.7% accuracy improvement, with a final peak accuracy of 93.6%. Full article

Aircraft targets in SAR imagery often exhibit discrete scattering characteristics, significant variations in pose and scale, strong speckle noise in background clutter, and complex background interference, which jointly hinder stable structural feature extraction and accurate target localization. Existing detectors for SAR aircraft recognition primarily rely on bounding-box regression and classification; they do not completely exploit target structural cues, spatial attention, and frequency-domain information. To address these limitations, we propose a collaborative detection framework that integrates an uncertainty-aware keypoint-driven module (UAKM) with a fractional Fourier convolution backbone (S-FRConv). UAKM introduces a center-keypoint regression branch that jointly predicts keypoint coordinates and Laplacian scale parameters and employs a 2D Laplace negative log-likelihood loss to estimate uncertainty. The derived dense uncertainty heatmap is then used as spatial attention weights to guide distribution-based regression and multi-scale feature re-weighting, without requiring any additional annotations. S-FRConv embeds the Fractional Fourier Transform into shallow backbone layers and C2f modules, enabling joint spatial–spectral feature modeling that suppresses speckle noise and enhances edge and orientation representations. Experiments on the public SAR-AIRcraft-1.0 dataset demonstrate that the proposed method systematically improves the detection performance. For the Nano model, the overall mAP50 increases from 0.810 to 0.867, and the mAP 50:95 improves from 0.637 to 0.655 compared with the baseline, corresponding to gains of 5.7 and 1.8 percentage points, respectively. These results validate the effectiveness and generalization potential of combining uncertainty-driven spatial attention with fractional spectral feature enhancement for SAR aircraft target detection. Full article

The selective detection of small, highly hydrophilic metabolites differing only in stereochemistry represents a major challenge in biosensor development. Here, we present a computational investigation to elucidate the molecular basis of the experimentally observed selectivity of a protein-based electrochemical biosensor toward myo-inositol over D-chiro-inositol. Although the two stereoisomers differ only in the orientation of a single hydroxyl group, they induce distinct dynamic effects on the protein recognition element. Molecular docking revealed comparable binding regions and similar affinity scores, indicating that selectivity does not arise from differences in binding site or docking energy. To investigate dynamic contributions, all-atom molecular dynamics simulations were performed in triplicate (3 × 100 ns) using the AMBER99SB force field and explicit TIP3P water. Trajectory analyses showed that myo-inositol forms a more persistent hydrogen bond network, resulting in reduced residue-level flexibility, more stable ligand–protein interactions, and enhanced local structural stabilization. Overall, these findings support a dynamic model of stereoselective recognition in which ligand-induced modulation of protein conformational ensembles, rather than static affinity, governs biosensor performance. This work highlights the value of molecular dynamics simulations in the rational design of biosensors targeting structurally similar analytes. Full article

This research investigates atmospheric neutron-induced single event effects (SEEs) on advanced artificial intelligence (AI) chips during natural environment operation. The RK3588 neural processing unit (NPU) is the evaluated target chip, and its SEE is assessed through a combination of irradiation testing and software fault injection. During the irradiation test, the chip was exposed to a spectrum neutron at the China Spallation Neutron Source. Upon reaching a cumulative fluence of 8.25 × 10⁹ n·cm², a total of 14,018 soft errors were detected, of which 99.97% manifested as variations in target recognition accuracy and network inference latency. Among these variations, both detrimental effects (reduced target recognition accuracy or prolonged network inference time) and beneficial effects (enhanced target recognition accuracy or shortened network inference time) caused by single event effects were observed. In addition, atmospheric neutron single event effects were found to cause NPU operation suspension and system crashes. Based on the irradiation test results, failure predictions for neural processing units in real-world environments were estimated, and mitigation recommendations were proposed. Furthermore, software fault injections were employed to conduct in-depth analysis of detected soft errors during irradiation testing. This research provides support and references for the reliable application of artificial intelligence chips in natural environments. Full article

Melanoma treatment has been transformed by immune checkpoint blockade, yet many patients still experience primary resistance, limited durability of response, or acquired resistance. These limitations underscore the need for additional targets that reflect melanoma biology while enabling new therapeutic strategies, particularly in biologically defined settings of immune escape such as checkpoint-resistant, HLA-low, dedifferentiated, or stress-adapted melanoma. The B7-H6/NKp30 axis has gained attention as a link between tumor cell stress, immune recognition, and therapy-related adaptation. B7-H6 (NCR3LG1), an inducible ligand for NKp30, has been detected in melanoma cell lines and tumor specimens, and soluble B7-H6 has been identified in a subset of patients. Membrane-bound B7-H6 may support NK-cell activation, whereas ligand shedding and accumulation of soluble B7-H6 may reduce effective antitumor recognition and promote immune evasion. Emerging evidence further suggests that B7-H6 expression may be linked to tumor-intrinsic programs relevant to melanoma cell survival, migration, and adaptation to therapeutic stress. However, B7-H6 is not yet a validated predictive biomarker or an established therapeutic target in melanoma, and current evidence remains limited by small melanoma-specific datasets, incomplete information on spatial and temporal heterogeneity, and the absence of melanoma-focused clinical validation. In this review, we examine the role of the B7-H6/NKp30 axis in immune surveillance, tumor escape, biomarker development, and therapeutic targeting, and discuss its translational potential in melanoma as an emerging but incompletely validated pathway that warrants focused investigation in melanoma states where conventional immune control is limited. Full article

Neonatal seizures represent the most common neurological emergency in the neonatal period and arise within a uniquely immature and highly dynamic brain. Their recognition is challenging due to frequent electroclinical dissociation, with many seizures remaining purely electrographic and therefore detectable only through continuous electroencephalogram (cEEG) monitoring. This narrative review provides an integrated and updated overview of neonatal seizures, bridging developmental neurobiology, diagnostic challenges, etiological classification, and therapeutic strategies. The immature brain is characterized by an imbalance between excitation and inhibition, transient network architectures, and activity-dependent developmental processes, all of which contribute to the distinct electroclinical features of neonatal seizures. cEEG remains essential for accurate diagnosis and quantification of seizure burden, which may influence outcome. Etiology represents the primary determinant of prognosis, with hypoxic–ischemic encephalopathy (HIE), stroke, and genetic disorders among the most frequent causes. Advances in genetic testing have improved diagnostic precision and enabled targeted therapies in selected cases, supporting a precision medicine approach. Several key findings emerge from the current evidence base: (i) the neonatal brain is a developmentally constrained system in which excitation–inhibition imbalance, transient circuits and immature long-range connectivity shape an electroclinically distinct seizure phenotype; (ii) cEEG is the gold standard for detection and quantification of seizure burden, since the majority of neonatal seizures are electrographic-only and bedside clinical recognition systematically underestimates true seizure burden; (iii) etiology—chiefly HIE, stroke, and genetic causes—remains the strongest determinant of outcome, while seizure burden acts as an independent and potentially modifiable prognostic modifier; (iv) phenobarbital retains an evidence-based advantage in acute electrographic seizure control, whereas levetiracetam offers a favorable safety profile in the absence of robust long-term human neurotoxicity data; (v) rapid genomic diagnostics, artificial intelligence-assisted EEG analysis and multimodal neuromonitoring are converging toward a precision-neonatology framework, but their translation into routine practice requires validation, standardization, and equitable access. Future neonatal seizure care should extend beyond seizure control to the preservation and optimization of neurodevelopmental outcomes. Full article

Robotic sorting-and-filling tasks in flexible manufacturing require robots to reproduce specified in-box arrangements while adapting to variations in container poses, object availability, sensing conditions, and external interventions. This paper proposes a box-centric one-shot teaching framework for robotic packing tasks with relative pose constraints. In the teaching stage, a human operator demonstrates the desired packing layout only once. The system uses reference-prompted SAM-based contour refinement to extract box and in-box object contours, object categories, quantities, and relative position and orientation constraints. These constraints are then converted from pixel-plane measurements into box-local pose constraints, forming a reusable box-centric packing template that preserves both translational and angular layout information. During execution, the recorded template is transferred to detected box instances with different global poses, and executable pick-and-place commands are generated through a task-level perception-to-command pipeline. A mechanism for continuous assignment and state updates is further introduced to maintain residual target slots, update object-to-slot allocation, and report missing or redundant objects across execution rounds. Single-box template transfer experiments achieved mean placement errors of 7.16 mm and 7.57 mm for two recorded templates, while representative post-execution images further showed that the relative object orientations were visually preserved with respect to the taught template footprints. Multi-box experiments demonstrated that unfinished residual slots could be preserved and completed after scene updates without re-teaching. Additional validation with different container types and object shapes showed the feasibility of extending the framework beyond cube-only cases. Ablation tests under nine exposure settings further showed that SAM refinement improved template-acquisition robustness compared with the previous recognition method. These results verify that the proposed framework enables one-shot template acquisition, box-centric layout transfer, relative pose preservation, and persistent task-level execution for constrained robotic packing tasks. Full article

With the rapid development of maritime transportation and resource development activities, maritime distress events are increasingly frequent, and efficient and accurate target recognition and rescue response methods are urgently needed. The traditional monitoring methods are limited by efficiency and real time, which is difficult to adapt to the complex and changeable marine environment. Therefore, based on the RT-DETR model of transformer architecture, an improved scheme for maritime distress target detection is proposed to improve the small target recognition ability and detection efficiency. Specific improvements include: a small target-focused convolution module (SFConv) is designed to enhance the efficiency of feature extraction and reasoning of small-scale targets; The cross-scale feature interaction optimization module (SPE) is further proposed to improve the ability of multi-scale perception and background suppression; The Focaler-DIoU loss function is introduced to enhance the discrimination performance of the model for difficult samples. On the basis of maintaining the end-to-end detection advantage of RT-DETR, the improvement is of 0.83474, which is 5.7% higher than the original model (0.78964). The accuracy and robustness of the model in complex marine environment is significantly improved, and technical support is provided for the construction of an efficient and intelligent marine monitoring and emergency response system. Full article