Search Results (28,786)

The expansion of Internet of Things (IoT) networks has enabled real-time data collection and automation across smart cities, healthcare, and agriculture, delivering greater convenience and efficiency; however, exposure to diverse threats has also increased. Machine learning-based Intrusion Detection Systems (IDSs) provide an effective means of defense, yet they require large volumes of data, and the use of raw IoT network data containing sensitive information introduces new privacy risks. This study proposes a novel privacy-preserving synthetic data generation model based on a tabular diffusion framework that incorporates Differential Privacy (DP). Among the three diffusion models (TabDDPM, TabSyn, and TabDiff), TabDiff with Utility-Preserving DP (UP-DP) achieved the best Synthetic Data Vault (SDV) Fidelity (0.98) and higher values on multiple statistical metrics, indicating improved utility. Furthermore, by employing the DisclosureProtection and attribute inference to infer and compare sensitive attributes on both real and synthetic datasets, we show that the proposed approach reduces privacy risk of the synthetic data. Additionally, a Membership Inference Attack (MIA) was also used for demonstration on models trained with both real and synthetic data. This approach decreases the risk of leaking patterns related to sensitive information, thereby enabling secure dataset sharing and analysis. Full article

In security applications, visible-light pedestrian detectors are highly sensitive to changes in illumination and fail under low-light or nighttime conditions, while infrared sensors, though resilient to lighting, often produce blurred object boundaries that hinder precise localization. To address these complementary limitations, we propose a practical multimodal pipeline—Adaptive Energy–Gradient–Contrast (EGC) Fusion with AIFI-YOLOv12—that first fuses infrared and low-light visible images using per-pixel weights derived from local energy, gradient magnitude and contrast measures, then detects pedestrians with an improved YOLOv12 backbone. The detector integrates an AIFI attention module at high semantic levels, replaces selected modules with A2C2f blocks to enhance cross-channel feature aggregation, and preserves P3–P5 outputs to improve small-object localization. We evaluate the complete pipeline on the LLVIP dataset and report Precision, Recall, mAP@50, mAP@50–95, GFLOPs, FPS and detection time, comparing against YOLOv8, YOLOv10–YOLOv12 baselines (n and s scales). Quantitative and qualitative results show that the proposed fusion restores complementary thermal and visible details and that the AIFI-enhanced detector yields more robust nighttime pedestrian detection while maintaining a competitive computational profile suitable for real-world security deployments. Full article

Selective de-identification is a key technology for protecting sensitive objects in visual data while preserving meaningful information. This study proposes a framework that leverages text prompt-based zeroshot and referring object detection techniques to accurately identify and selectively de-identify sensitive objects without relying on predefined classes. By utilizing state-of-the-art models such as GroundingDINO, objects are detected based on natural language prompts, and de-identification—via blurring or masking—is applied only to the corresponding regions, thereby minimizing information loss while achieving a high level of privacy protection. Experimental results demonstrate that the proposed method outperforms conventional batch de-identification approaches in terms of scalability and flexibility. Full article

Background/Objectives: Bangkeehwangkee-tang (BHT) is a traditional herbal formula composed of six medicinal herbs: Sinomenii Caulis et Rhizoma, Astragali Radix, Atractylodis Rhizoma Alba, Zingiberis Rhizoma Recens, Zizyphi Fructus, and Glycyrrhizae Radix et Rhizoma. BHT has been widely used for its immunomodulatory and anti-inflammatory effects. This study aimed to develop a reliable analytical method for the simultaneous determination of 22 marker compounds to ensure consistent quality control and to ensure consistent efficacy in both clinical and non-clinical studies of BHT. Methods: An ultra-performance liquid chromatography–tandem mass spectrometry (UPLC–MS/MS) method based on multiple reaction monitoring was developed and validated for the simultaneous determination of 22 marker compounds in BHT. The method was evaluated for selectivity, linearity (coefficient of determination, r²), sensitivity (limit of detection (LOD) and limit of quantification (LOQ)), accuracy (recovery), and precision (relative standard deviation (RSD)) in accordance with guidelines. Results: The developed method exhibited excellent selectivity and linearity (r² ≥ 0.9913) for all target compounds. The LOD and LOQ ranged from 0.09 μg/L to 326.58 μg/L and 0.28 μg/L to 979.75 μg/L, respectively. The recovery ranged from 90.36% to 113.74%, and precision (RSD) was ≤15%, confirming the method’s reliability. The application of the method to various BHT samples revealed substantial variations in the marker compound contents, particularly for sinomenine, magnoflorine, and glycyrrhizin. Conclusions: These findings highlight the necessity for standardized quality control of BHT and demonstrate that the developed UPLC–MS/MS method is a practical and reliable tool for performing quality assessment of traditional herbal formulas. Full article

Pythiosis is an underestimated and neglected disease in Brazil, both in horse breeders and in horses. The molecular detection of P. insidiosum in horses in the Brazilian Northeast represents a milestone in the epidemiology of equine pythiosis in the country. This study reports novel cases of equine pythiosis, diagnosed by molecular methods, in five states of Northeastern Brazil. Clinical samples were submitted to microbiological culture, DNA extraction, and nested-PCR for molecular detection of P. insidiosum. The nested-PCR successfully detected P. insidiosum in four out of five equine lesion samples, demonstrating higher sensitivity compared to microbiological culture, which confirmed the pathogen in only one case. These findings reinforce the effectiveness of molecular diagnosis for equine pythiosis compared to conventional methods. This is the first molecular confirmation of P. insidiosum in horses from Northeastern Brazil, emphasizing the need for broader surveillance and improved diagnostic approaches in neglected regions. Full article

Background: B7-H3, an immunoregulatory protein of the B7 family, has been associated with both anti-cancer immunity and tumor promotion, with its expression commonly correlated with poor prognosis. Although it is frequently expressed across cancers, its heterogeneity may limit the effectiveness of B7-H3-targeted therapies. Consequently, a sensitive and non-invasive method is needed to assess B7-H3 expression for patient selection and stratification. Single-domain antibody fragments (sdAbs) offer a promising platform for developing such a diagnostic tool. Methods: To generate B7-H3 sdAbs, two Ilamas were immunized with the recombinant human B7-H3 protein. Positive clones were selected through Phage biopanning and characterized for thermal stability, binding specificity, and affinity to human and murine B7-H3 proteins. Selected sdAbs were radiolabeled with Technetium-99m (^99mTc) and evaluated for B7-H3 detection in two xenograft tumor models using micro-SPECT/CT imaging and dissection studies. Results: Sixteen purified sdAbs bound specifically to recombinant B7-H3 proteins and cells expressing native B7-H3 antigens, with nanomolar affinities. The four best-performing sdAbs bound promiscuously to tested mouse and human B7-H3 isoforms. Lead sdAb C51 labeled with ^99mTc displayed specific accumulation across two human B7-H3⁺ tumor models, achieving high contrast with a tumor-to-blood ratio of up to 10 ± 3.16, and a tumor uptake of up to 4.96 ± 1.4%IA/g at 1.5 h post injection. Conclusions: The lead sdAb enabled rapid, specific, and non-invasive imaging of human B7-H3⁺ tumors. Its isoform promiscuity supports broad applicability across cancers expressing different human B7-H3 isoforms. These results support further development for clinical translation to enable patient selection and improved B7-H3-targeted therapies. Full article

Flexible and wearable sensors have emerged as transformative technologies in the field of real-time health monitoring, offering non-invasive, continuous, and personalized healthcare solutions. These devices are designed to conform intimately to the human body, enabling seamless detection of vital physiological and biochemical signals under dynamic conditions. Recent advancements in material science and device engineering have led to the development of sensors with enhanced sensitivity, biocompatibility, and wearability, addressing the growing demand for preventive healthcare and remote patient monitoring. This review provides a comprehensive overview of the progress in flexible wearable sensors, including novel materials, sensor designs, and system integration strategies. It begins by surveying the latest advances in substrate and functional materials and hybrid structures that enable mechanical flexibility, skin conformability, and high sensitivity. The review then examines various sensor mechanisms and their implementation in monitoring vital signs, physical activity, and chronic diseases. Real-world applications are explored in depth, covering scenarios from cardiovascular and respiratory monitoring to motion tracking and rehabilitation support. Despite the significant strides made, challenges related to material robustness, sensor accuracy, and multi-modal integration remain, and this review discusses these challenges alongside potential future directions for enhancing the functionality and adoption of flexible wearable sensor systems. Full article

Lipopolysaccharides (LPS) from Gram-negative bacteria represent a significant challenge across various industries due to their prevalence and pathogenicity and the limitations of existing detection methods. Traditional approaches, such as the rabbit pyrogen test (RPT) and the Limulus Amebocyte Lysate (LAL) assay, have served as gold standards for endotoxin detection. However, these methods are constrained by high costs, lengthy processing times, environmental concerns, and the need for significant reagent volumes, which limit their scalability and application in resource-limited settings. In this study, we introduce an innovative microfluidic platform that integrates the LAL assay within microdroplets, addressing the critical limitations of traditional techniques. By leveraging the precise fluid control and reaction isolation offered by microdroplet technology, the system reduces reagent consumption, enhances sensitivity, and enables high-throughput analysis. Calibration tests were performed to validate the platform’s ability to detect LPS, using colorimetric measurements. Results demonstrated comparable or improved performance relative to traditional systems, achieving lower detection limits and greater accuracy. This work demonstrates a proof-of-concept miniaturisation of the pharmacopoeial LAL assay. The method yielded low intra-assay variability (σ ≈ 0.002 OD; CV ≈ 0.9% over n = 50 droplets per point) and a LOD estimated from calibration statistics after path-length normalisation. Broader adoption will require additional comparative validation and standardisation. This scalable, cost-effective, and environmentally sustainable approach offers a practical solution for endotoxin detection in clinical diagnostics, biopharmaceutical production, and environmental monitoring. The proposed technology paves the way for advanced LPS detection methods that meet stringent safety standards while improving efficiency, affordability, and adaptability for diverse applications. Full article

Conventional diagnosis of dermatophytosis relies on fungal culture and microscopic examination, methods that are often time-consuming and lack sensitivity. This study aimed to develop and compare real-time PCR assays for the simultaneous detection and differentiation of three major dermatophytes in dogs: Microsporum canis, Nannizzia gypsea, and Trichophyton mentagrophytes. Three qPCR platforms targeting the chitin synthase 1 (CHS1) gene—SYBR Green, EvaGreen, and dual-emission fluorescence resonance energy transfer (FRET)—were evaluated. The FRET assay demonstrated the highest performance, achieving a detection limit of a single gene copy per reaction and producing distinct melting profiles that enabled accurate species discrimination (M. canis ~56.1 °C, N. gypsea ~53.0 °C, T. mentagrophytes ~51.8 °C). In contrast, SYBR Green and EvaGreen assays showed reduced sensitivity and cross-reactivity with non-target fungi. All assays were validated using three ATCC reference strains, ten clinical isolates of the target dermatophytes, and nine additional fungal species, including Nocardia, Aspergillus, Fusarium, Sporothrix, and Candida. Overall, FRET-qPCR exhibited a 100% specificity and a detection limit of one copy of target gene per reaction, offering a rapid, reliable tool for accurate diagnosis and molecular surveillance of dermatophytosis in companion animals. Full article

The germination, seedling growth, and crop productivity of maize seeds are significantly impacted by seed aging. This study investigated the efficacy of silver nanoparticles (AgNPs) as a seed priming agent for maize inbred lines exhibiting varying degrees of aging tolerance. Two inbred lines, aging-sensitive I178 and aging-tolerant X178, were used. AgNP treatment significantly promoted the germination of I178 (from 55% to 85%, compared with water treatment). Notable improvements were observed in root length, shoot length, and lateral root formation after AgNP treatment in I178. However, X178 showed no significant changes in germination and seedling growth after the AgNP treatment. Further transcriptomic analysis was performed on X178 and I178 before (water treatment) and after AgNP treatment to study genes and the expression network of the mechanism induced by AgNP promotion. In I178, AgNP treatment led to a substantial increase in differentially expressed genes (DEGs). A total of 800 DEGs were identified, with 517 being upregulated and 283 downregulated. The DEGs in I178 were mainly involved in metabolic processes, stress responses, and membrane repair. For example, genes related to lipid metabolism and membrane integrity were upregulated, along with seven genes associated with antioxidant action and redox metabolism. This indicates that AgNPs might enhance membrane stability and stress tolerance in I178. In contrast, X178 had a limited transcriptomic response to AgNP treatment. Although 874 DEGs were detected, the number of genes related to key processes like those in I178 did not change significantly, which is in line with its inherent aging tolerance. Overall, these results suggest that AgNPs can effectively improve seed vigor and counteract the negative effects of seed aging, especially in aging-sensitive maize lines. The mechanism likely occurs through regulating gene expression related to stress response, metabolic repair, and membrane stability. This study provides new insights into the molecular basis of AgNP-mediated seed vigor enhancement, which has potential implications for improving seed quality in agricultural production. Full article

Personalized, point-of-care testing of human tears is essential for ocular disease diagnosis, yet it is hampered by picomolar biomarker levels and microliter sample volumes. In this work, we developed an integrated, portable electrochemiluminescence (ECL)-based device for rapid and quantitative detection of tumor necrosis factor alpha (TNF-α), a pivotal inflammatory marker in ocular surface disease, with particular relevance to dry eye syndrome (DES). The device integrates a miniaturized electrochemical cell for ECL reactions and a compact silica photomultiplier for signal measurement. A vertical silica mesochannel (VSM)-coated ITO electrode is also integrated and further functionalized with TNF-α-specific aptamers. The VSM enables the enrichment of ECL luminophores, thus enabling further amplification of ECL signals and enhancing sensitivity. A wide linear range from 0.1 to 200 pg/mL was achieved using 10-fold dilution of 3 μL tear samples. Overall, this study provides a portable, highly sensitive platform for personalized analysis of TNF-α in tear fluid, enabling rapid point-of-care assessment of DES. Full article

Background/Objectives: The global incidence of thyroid cancer has been increasing, necessitating improved diagnostic strategies for detecting lymph node metastases. Fine-Needle Aspiration Biopsy (FNA) is a widely used diagnostic tool; however, its accuracy is sometimes limited, particularly in cases with non-diagnostic results. Washout Thyroglobulin (Washout Tg) measurement has emerged as an important adjunctive tool in refining thyroid cancer diagnosis. Methods: This retrospective study analyzed 723 patients who underwent thyroid cancer surgery at Samsung Medical Center from 2013 to 2023. The patients were categorized based on their thyroid status into three groups: 1. total thyroidectomy with modified radical neck dissection (mRND); 2. completion thyroidectomy with mRND; and 3. mRND or selective neck dissection (SND) without thyroidectomy. The Washout Tg levels and their diagnostic performance were evaluated using Receiver Operating Characteristic (ROC) analysis, determining the optimal cutoff values for predicting lymph node metastasis. Results: Washout Tg demonstrated high sensitivity for detecting metastases, with the optimal cutoff values varying based on thyroid status. For the patients who had total thyroidectomy, the cutoff was 23.3 ng/mL (AUC = 0.85, sensitivity = 82.6%, and specificity = 75.0%). In completion thyroidectomy cases, a threshold of 7.2 ng/mL (AUC = 0.879) achieved 98.4% sensitivity and 80.0% specificity. For patients without thyroidectomy (mRND/SND group), a cutoff of 0.1 ng/mL (AUC = 0.766) yielded 98.9% sensitivity but lower specificity (60.0%). Additionally, the Washout Tg/serum Tg ratio demonstrated high diagnostic accuracy with a cutoff of >1 (sensitivity = 97.09% and specificity = 63.64%). Conclusions: The Washout Tg measurement and the Washout Tg/serum Tg ratio play a crucial role in detecting lymph node metastases, particularly in patients post-thyroidectomy. The findings emphasize the necessity of adjusting the Washout Tg cutoff values based on thyroid status to enhance diagnostic accuracy. Further prospective studies are required to validate these thresholds and optimize their clinical application. Full article

Previous studies on Sound Event Localization and Detection (SELD) have primarily focused on CNN- and Transformer-based designs. While CNNs possess local receptive fields, making it difficult to capture global dependencies over long sequences, Transformers excel at modeling long-range dependencies but have limited sensitivity to local time–frequency features. Recently, the VMamba architecture, built upon the Visual State Space (VSS) model, has shown great promise in handling long sequences, yet it remains limited in modeling local spatial details. To address this issue, we propose a novel state space model with an attention-enhanced feature fusion mechanism, termed FFMamba, which balances both local spatial modeling and long-range dependency capture. At a fine-grained level, we design two key modules: the Multi-Scale Fusion Visual State Space (MSFVSS) module and the Wavelet Transform-Enhanced Downsampling (WTED) module. Specifically, the MSFVSS module integrates a Multi-Scale Fusion (MSF) component into the VSS framework, enhancing its ability to capture both long-range temporal dependencies and detailed local spatial information. Meanwhile, the WTED module employs a dual-branch design to fuse spatial and frequency domain features, improving the richness of feature representations. Comparative experiments were conducted on the DCASE2021 Task 3 and DCASE2022 Task 3 datasets. The results demonstrate that the proposed FFMamba model outperforms recent approaches in capturing long-range temporal dependencies and effectively integrating multi-scale audio features. In addition, ablation studies confirmed the effectiveness of the MSFVSS and WTED modules. Full article

Background: Gastric cancer (GC) is the fifth most common malignancy worldwide. Early detection of precancerous conditions—atrophic gastritis (AG), intestinal metaplasia (IM), and dysplasia—is vital for surveillance. Objectives: To assess the accuracy of single high-quality endoscopy (HQE) in detecting advanced GPCs and to identify risk factors for AG, IM, and dysplasia. Methods: A retrospective review of 442 gastroscopies (2017–2022) at a single center. Endoscopic findings were compared with histology, including OLGA/OLGIM staging, dysplasia, and Helicobacter pylori (H. pylori) status. Results: The study population comprised 319 women (72.17%) and 123 men (27.83%), with a mean age of 59 years (SD: 12.53). AG, as defined by OLGA and OLGIM staging, was identified in 90 patients (20.36%) and 50 patients (11.31%), respectively. A total of 44 cases of de novo gastric dysplasia were observed, while HP infection was confirmed in 37 individuals (8.37%). We observed similar low sensitivity for detection of advanced OLGA (32.5%), OLGIM (40%), and dysplasia (19.7%) with relatively high specificity (~89%). Advanced AG and IM peaked at ages 51–53. Risk factors for advanced OLGIM included male sex (OR 2.26; p < 0.001) and presence of dysplasia (OR 2.09; p = 0.02). Dysplasia was positively associated with AG (OR 2.03; p < 0.001) and IM (OR 2.21; p < 0.001) but inversely associated with a family history of GC (OR 0.44; p < 0.001). Conclusions: A single HQE can help exclude advanced GPCs, but due to low sensitivity, gastric mapping biopsies remain crucial. Males are at increased risk of extensive IM. Family history of GC was linked to lower OLGA/OLGIM stages. Full article

This study presents a comparative evaluation of post-process sensor integration in additively manufactured (AM) metal and the in-situ process for polymer structures for structural health monitoring (SHM), with an emphasis on embedded sensors. Geometrically identical specimens were fabricated using copper via metal fused filament fabrication (FFF) and PLA via polymer FFF, with piezoelectric transducers (PZTs) inserted into internal cavities to assess the influence of material and placement on sensing fidelity. Mechanical testing under compressive and point loads generated signals that were transformed into time–frequency spectrograms using a Short-Time Fourier Transform (STFT) framework. An engineered RGB representation was developed, combining global amplitude scaling with an amplitude-envelope encoding to enhance contrast and highlight subtle wave features. These spectrograms served as inputs to convolutional neural networks (CNNs) for classification of load conditions and detection of damage-related features. Results showed reliable recognition in both copper and PLA specimens, with CNN classification accuracies exceeding 95%. Embedded PZTs were especially effective in PLA, where signal damping and environmental sensitivity often hinder surface-mounted sensors. This work demonstrates the advantages of embedded sensing in AM structures, particularly when paired with spectrogram-based feature engineering and CNN modeling, advancing real-time SHM for aerospace, energy, and defense applications. Full article