Search Results (947)

Conventional terahertz (THz) radio frequency (RF) frontends struggle to simultaneously balance the high performance and miniaturization of monolithic integrated designs with the excellent testability of discrete modular architectures. This paper presents a detachable 183 GHz terahertz RF frontend and completes the module design and system integration of a low-noise amplifier (LNA) and a second-order subharmonic mixer. Through optimization of the waveguide-to-microstrip transition, parasitic compensation for pad bonding, and the structural design of the chip shielding cavity, combined with a high-precision alignment scheme using positioning pins and screws, the integrated module achieves detachability, testability, and ease of maintenance. Measurement results show that across the 160–200 GHz frequency band, the amplifier achieves an average gain of 16.51 dB; the mixer exhibits a minimum conversion loss of 8.62 dB; and the full-link noise figure of the system reaches 6.68 dB. The proposed scheme effectively addresses the engineering challenges of conventional integrated architectures and provides a practical implementation pathway for terahertz communication and remote sensing detection frontends. Full article

Soybean rust caused by the fungus Phakopsora pachyrhizi threatens global soybean production, causing yield losses of up to 80%. Race-specific Rpp genes provide short-term resistance due to pathogen variability, whereas partial resistance (PR) offers durable, broad-spectrum protection, though its genetic basis remains unclear. This study aimed to identify genomic regions and candidate genes underlying PR using the Fixed and Random Model Circulating Probability Unification (FarmCPU) genome-wide association study (GWAS) and machine learning (ML) methods, Random Forest (RF) and Support Vector Regression (SVR). A panel of 312 soybean accessions was evaluated under natural infection across six Ugandan environments. Rust index (RI), derived from rust severity and sporulation level, was used to estimate heritability (H²) and rank genotypes through Best Linear Unbiased Predictions (BLUPs), while Best Linear Unbiased Estimators (BLUEs) supported GWAS input. After quality control, 8272 SNPs were analyzed within a ±60 kb linkage disequilibrium (LD) window. Multi-environmental Analysis (MEA) of RI showed significant genetic effects (p < 0.01); H² = 0.57–0.68. Sixty-one loci were detected: six by FarmCPU, 15 by RF, and 41 by SVR. Key genes included Glyma.01G128100 (a WRKY transcription factor) and Glyma. 13G228000, receptor-like kinase) and Glyma.20G173100 (WD40-domain regulator). Integrating ML with GWAS improved locus detection, confirming the polygenic nature of PR and supporting the use of genomic selection and locus pyramiding for durable rust resistance. Full article

Background: Weight loss and skeletal muscle wasting are frequent in cancer and may influence treatment tolerance and outcomes. Computed tomography (CT) based body composition analysis at the third lumbar vertebra (L3) is an accurate method to quantify skeletal muscle in routine oncology care. Methods: We performed a multicenter retrospective cohort study including 202 adults with locally advanced (stage IB–III) gastric cancer treated in four Portuguese hospitals (January 2020–December 2022). Skeletal muscle area (SMA) was assessed on baseline CT at the L3 vertebral level, using Data Analysis Facilitation Suite (DAFS) software v3.11.2, and skeletal muscle index (SMI) was subsequently calculated. Patients with low muscle quantity were classified as sarcopenic (below sex-specific SMI mean). We evaluated associations with relapse-free survival (RFS), overall survival (OS), FLOT chemotherapy dose-limiting toxicities (DLTs), and postoperative complications after gastrectomy. Results: Mean age was 69 years, 65% had ECOG PS 0, 53% received FLOT chemotherapy protocol. Mean SMI was 49.6 cm²/m² in males and 40.9 cm²/m² in females and correlated positively, though moderately, with BMI (p < 0.01; r = 0.424). Sarcopenia was not significantly associated with RFS (p = 0.186) or OS (p = 0.168) at 30-month follow-up. Although numerical differences were observed (64% vs. 56% of patients did not relapse and 74% vs. 63% were alive, for non-sarcopenic vs. sarcopenic patients). Sarcopenia was associated with a higher risk of DLTs (p = 0.021; OR 2.56, 95% CI 1.15–5.73) and postoperative complications (p = 0.024; OR 2.16, 95% CI 1.11–4.21). Conclusions: Sarcopenia significantly increases the risk of chemotherapy toxicity and postoperative complications in locally advanced gastric cancer. However, its effect on OS and RFS was not statistically significant at 30-month follow-up. Standardization of CT-based sarcopenia cut-offs remains a major barrier to clinical implementation. Full article

Background/Objectives: Cancer cachexia (CC) is a metabolic syndrome characterized by the progressive loss of skeletal muscle and adipose tissue during tumor progression. Despite its clinical prevalence, effective therapeutic options are currently lacking. Phloretin, a natural flavonoid with potent anti-inflammatory and antioxidant properties, has unclear efficacy against CC. This study investigates the therapeutic potential of phloretin in ameliorating cancer cachexia. Methods: Mouse models of CC were established using BALB/c mice implanted with C26 colon carcinoma cells and C57BL/6 mice implanted with Lewis lung carcinoma (LLC) cells. Upon the detection of palpable tumors, phloretin (10 mg/kg) was administered daily via intraperitoneal injection. At the endpoint, hind limb skeletal muscle, inguinal white adipose tissue (iWAT), and hearts were harvested and weighed. Lean body mass was assessed by analyzing the weight of the carcass following the excision of skin, subcutaneous fat, and visceral organs. Gene expression and protein levels in muscle tissues were subsequently quantified. Results: Phloretin administration significantly alleviated tumor-induced loss of tumor-free body weight. It effectively preserved skeletal muscle mass in both C26 and LLC cachexia models, while significantly attenuating adipose tissue depletion in the C26 model. In vitro, phloretin treatment mitigated myotube atrophy induced by C26 conditioned medium. Mechanistically, phloretin inhibited STAT3 activation in skeletal muscle. This inhibition suppressed the expression of the E3 ubiquitin ligases MuRF-1 and Atrogin-1. Furthermore, phloretin concurrently modulated the autophagy pathway. Conclusions: Phloretin effectively ameliorates cancer cachexia-induced muscle wasting by targeting STAT3-mediated protein degradation and autophagy pathways. These findings suggest that phloretin represents a promising therapeutic agent for the clinical management of cancer-associated cachexia. Full article

The redox mechanisms of RAW 264.7 macrophages exposed to 2.45 GHz RF-EMF at subthermal specific absorption rates and to lipopolysaccharide (LPS) and/or the SARS-CoV-2 spike protein (CSP) were investigated. To this end, cellular responses (lysosomal and mitochondrial activity, nitric oxide (NO) production, and cell survival/death) were measured after 6, 24, and 48 h. Selective loss of viability in cells exposed to RF and LPS was observed at 6 h, consistent with early defects in membrane permeability. Lysosomal activity was significantly enhanced in cells treated with RF + LPS. Mitochondrial activity decreased in cells exposed to RF + LPS at 6 h and increased in cells treated with RF + CPS/LPS. Cell viability decreased greatly in cells treated with LPS and CSP + LPS after 24, particularly after 48 h. Nitrite levels peaked in non-irradiated cells treated with RF + LPS and in CSP + LPS at 24 h and decreased in irradiated cells after 48 h. Irradiation affected selection of the death mode: apoptosis decreased or remained unchanged in cells subjected to any of the treatments, while necrosis increased in cells treated with CPS, LPS, or both for 48 h. The combination of RF-EMF and infectious agents reprogrammed the interaction between mitochondria/lysosomes/nitric oxide (NO)/cell death in macrophages in a time- and stimulus-dependent manner. Full article

Frequency response analysis (FRA) is recognized as an effective method in power transformer winding fault diagnosis. However, the traditional numerical index methods focus on the overall features of FRA curves, making it difficult to capture subtle deformations in transformer windings. Similarly, existing digital image processing methods rely on a single feature or a simple feature combination without adaptive fusion. These methods ignore differences in the data distributions of features, leading to feature mismatch, the loss of sensitive fault information, and lower diagnostic accuracy. To solve this problem, a novel adaptive multiple-image-feature fusion method for transformer winding fault diagnosis is proposed. First, a multi-dimensional feature space combining image pixel matrix similarity, morphological features, and image texture features is built to decode the difference in fault of FRA images. Second, the multiple kernel learning (MKL) framework is used to dynamically adjust the fusion weights of different kernels to make features compatible and remove redundant information. Finally, comparative and ablation experiments show that the proposed method outperforms the traditional methods in identifying different types and levels of faults. The method achieves over 99% accuracy in fault type identification across SVM, KNN, and RF classifiers. For radial deformation (RD) severity prediction, the accuracy of the proposed model is 93.37% with SVM and 94.85% with KNN, outperforming the full-feature concatenation method. These results confirm the method’s robustness and diagnostic precision. Full article

Despite its continued growth, cloud computing remains susceptible to significant security challenges, as shared virtualised environments pose threats at multiple levels. These vulnerabilities are caused by a lack of security coverage in the responsibility model between the provider and the tenant. In this work, we propose the multi-layered architecture VIRTUOSO (VIRTual Unified Operation Security Optimiser) to cover these security gaps through advanced automation and ML. VIRTUOSO has four layers. The Input Layer extracts key risk components from collected telemetry data. The Deep Automation Security Layer provides automated actions and continuous monitoring of security defences. Its counterpart, the Intelligent Security Layer, predicts threats using anomaly detection. The last layer, the Output Layer, returns an aggregated risk summary. The datasets we used were chosen for their relevance: the UNSW-NB15 dataset, a subset of the web-attack classification from CSE-CIC-IDS2018, and a sample of anonymised log events from AWS CloudTrail. Our ensemble classifiers achieve a best accuracy of 95.08% ± 0.13% on UNSW-NB15 (RF), with statistically significant differences among models confirmed by the Friedman test (p < 0.004) and Nemenyi post hoc analysis, and 99.25% ± 0.52% on web-attack (CatBoost), where ensemble differences are not statistically significant (p = 0.093), consistent with the high separability of this dataset. The training-test gap and DNN curves show no overfitting, whereas our adversarial tests show a maximum accuracy loss of 8.1% at ε = 0.02. With these promising results, we can assert that, pending verification in an actual cloud environment and potential integration with FL, our ensemble classifier model appears to be a good real-world prototype. Full article

We propose a monolithically integrated optical frequency comb (OFC) generation platform on thin-film lithium niobate (TFLN), featuring cascaded dual-drive Mach–Zehnder modulators (DDMZM) and a ${\mathrm{S}\mathrm{i}}_3{\mathrm{N}}_4$-assisted spot size converter (SSC). To capture microscopic mode mismatches and spatial phase accumulation often overlooked in idealized scalar simulations, we establish a multi-physics co-simulation framework integrating finite-difference time-domain (FDTD) analysis with macroscopic transmission modeling. Based on this framework, the cascaded modulator architecture generates 25 highly stable comb lines with a dense 2 GHz spacing and an envelope flatness within 2 dB. Tolerance analysis indicates that the comb generation is highly resilient to typical manufacturing and environmental variations, including thermal bias drift, RF phase mismatch, and half-wave voltage ($V_{\pi }$) dispersion. Furthermore, physical-layer modeling shows that the integrated SSC reduces fiber-to-chip coupling loss to 0.55 dB per facet, preserving the necessary optical power budget. To validate the platform’s viability as a multi-wavelength continuous-wave source for spatial-division multiplexed (SDM) interconnects, a parallel transmission over a 20 km standard single-mode fiber is modeled. Using a digital signal processing (DSP)-free 10 Gb/s non-return-to-zero (NRZ) scheme, the 25-channel system maintains a worst-case bit error rate strictly below the forward error correction (FEC) threshold. This work offers a practical, physics-based evaluation framework for high-density co-packaged optics (CPO). Full article

This paper presents a 13 W Ku-band GaN HEMT MMIC power amplifier employing a coupled-line interstage stabilization technique for radar sensor front-end applications. High-efficiency and stable power amplification in the Ku-band is essential for radar sensing systems, where low-frequency instability and process sensitivity often limit multistage GaN amplifier performance. To address these challenges, a coupled-line interstage network is introduced instead of conventional series capacitors and parallel RC stabilization circuits. The proposed structure effectively suppresses low-frequency gain while maintaining RF performance and improving robustness against process variations due to its planar transmission-line implementation. The two-stage power amplifier was fabricated using a 0.25 μm commercial GaN HEMT MMIC process. For compact implementation, the coupled-line structure was realized in a meandered layout and verified through full electromagnetic simulations. Measured small-signal results show a gain (S21) of 18.6–21.6 dB, with input and output return losses (S11 and S22) of −3.3 to −10.2 dB and −4.4 to −7.2 dB, respectively, over 13.5–16 GHz. Large-signal measurements demonstrate a saturated output power of 40.7–41.5 dBm and a power-added efficiency of 21.3–28.1% across the same frequency range. The fabricated MMIC achieved stable operation without oscillation, validating the effectiveness of the proposed coupled-line stabilization approach for Ku-band radar sensor systems. Full article

Atmospheric corrosion of carbon and low-alloy steels causes direct economic losses that are estimated at around 3.4% of the global GDP, and its accurate multi-year prediction is essential for protective coating selection, service-life estimation, and infrastructure maintenance scheduling. In this study, machine learning (ML) algorithms, including gradient boosting regressor (GBR), eXtreme gradient boosting (XGBoost), random forest (RF), support vector regression (SVR), and ridge regression, were trained on a 600-sample physics-grounded dataset to predict the cumulative atmospheric corrosion loss (µm) of low-alloy steels over 1–10 years of exposure. The dataset was constructed using the exact ISO 9223:2012 dose–response function (DRF) for a first-year corrosion rate and the ISO 9224:2012 power-law multi-year kinetic model (C(t) = C₁·t0.5), spanning ISO 9223 corrosivity categories C2–CX across 11 environmental and material input features. All models were evaluated on the original (untransformed) corrosion scale under an 80/20 train/test split and five-fold cross-validation. Gradient boosting achieved the best overall performance with test set R² = 0.968, CV-R² = 0.969, RMSE = 10.58 µm, MAE = 5.99 µm, and MAPE = 12.6%. XGBoost was a close second (R² = 0.958, CV-R² = 0.960). RF achieved an R² of 0.944. SHAP (SHapley Additive exPlanations) analysis identified SO₂ deposition rate, exposure time, relative humidity, Cl⁻ deposition rate, and temperature as the five most influential predictors. The dominance of the SO₂ deposition rate (mean |SHAP| = 26.37 µm) and the high second-place ranking of exposure time (13.67 µm) are fully consistent with the ISO 9223:2012 dose–response function and ISO 9224:2012 power-law kinetics, respectively, while among the material features, Cu and Cr contents showed the strongest negative SHAP contributions, confirming their corrosion-inhibiting roles in weathering steels. These results establish a physics-consistent, interpretable ML benchmark exceeding R² = 0.90 for multi-year cumulative corrosion loss prediction and provide a quantitative tool for alloy screening, coating selection in aggressive atmospheric environments, and service-life planning. Full article

Cardiovascular disease (CVD) risk assessment in older adults requires models that are accurate, clinically interpretable, and able to retain performance in independent populations. This study developed an interpretable machine-learning framework for CVD risk stratification in individuals aged 65 years and older using routinely available clinical factors and a selected biochemical extension and then evaluated its performance in a substantially larger independent external cohort. Model development used a development cohort of 100 patients (Almaty, age ≥ 65) with leakage-free nested cross-validation and out-of-fold (OOF) probabilities. Three internally evaluated configurations were compared: a clinical logistic regression baseline (LR clinical), a biomarker-augmented logistic regression (LR selected), and a nonlinear random forest on the selected feature set (RF selected). Discrimination was assessed using ROC-AUC and PR-AUC; probabilistic accuracy using Brier score and log loss. Calibration was examined using OOF calibration curves with sigmoid calibration for selected models. Decision-analytic utility and exploratory operational thresholds were assessed using Decision Curve Analysis (DCA), yielding a three-tier scale with thresholds t__low = 0.23 and t__high = 0.40. In nested cross-validation, LR clinical achieved ROC-AUC 0.9425 ± 0.0188 and PR-AUC 0.9574 ± 0.0092 with Brier 0.1004 ± 0.0215 and log loss 0.3634 ± 0.0652; LR selected performed worse, while RF selected showed competitive discrimination. External validation on an independent cohort (n = 695) showed retained discrimination (ROC-AUC 0.8355; PR-AUC 0.9376) with acceptable probabilistic accuracy (Brier 0.1131; log loss 0.3760), and recalibration (intercept + slope) slightly improved probability metrics. Explainability analyses (odds ratios, permutation importance, SHAP) consistently identified heredity, BMI, physical activity, and diabetes as influential model-associated factors, with clinically plausible directionality. The results suggest that an interpretable model trained on a small geriatric cohort can retain meaningful predictive performance on a substantially larger external cohort, supporting the potential value of transparent risk stratification in older adults, while broader prospective and multi-center validation remains necessary before routine clinical implementation. Full article

This work presents the design and characterization of a thermoregulated, bandwidth-enhanced TEM cell system optimized for bioelectromagnetic experiments on biological cells, with a focus on bioluminescence resonance energy transfer investigations at 700 MHz and 3.5 GHz. Bandwidth improvement, achieved through geometric modifications and optimized connector transitions, resulted in reduced return and insertion losses and improved field uniformity, particularly in the 2.5–6 GHz range. Numerical simulations showed homogeneous electric field and normalized specific absorption rate (SAR) distributions (~1 W/kg) at 700 MHz. At 3.5 GHz, the improved TEM cell provided the most uniform exposure of the biological sample with SAR values of 15 W/kg and 10.5 W/kg, for the bulk and surface (bottom layer), respectively. Experimental SAR measurements using a ~1 mm³ fluoro-optic probe agreed well with simulations. To counteract RF-induced heating, the system incorporated active thermoregulation at 37 °C. At 3.5 GHz and 20 W input power, a 1.5 °C rise over 120 s was effectively mitigated using water-circulation cooling. This work provides a controlled and reliable setup for future studies on the interaction of 5G-band electromagnetic fields with biological systems. Full article

Rapid urban expansion and the growing spatial requirements of utility-scale photovoltaic (PV) deployment compete for the same category of land—flat, accessible, and high-insolation terrain—yet the scale, trajectory, and planning-sensitivity of this conflict remain poorly characterised at the regional level. This study quantifies the spatiotemporal competition between urban construction land and PV-suitable land in the Yangtze River Delta (YRD) from 2000 to 2020 and projects its evolution to 2030 under three development scenarios. Built-up areas were extracted for three epochs using a Random Forest (RF) classifier on the Google Earth Engine (GEE) platform, achieving overall accuracies of 87.7–94.5% and Kappa coefficients of 0.718–0.739. PV site suitability was evaluated through a hybrid Multi-Criteria Decision Analysis (MCDA) framework combining Boolean exclusion constraints with an Analytic Hierarchy Process (AHP)-based Weighted Linear Combination model; the weight structure was validated by a Consistency Ratio of 0.006, and a One-At-a-Time sensitivity analysis confirmed spatial robustness across threshold scenarios. Spatial overlay analysis reveals that the cumulative area of PV-suitable land occupied by urban built-up uses grew from 15,862 km² in 2000 to 23,872 km² in 2020, representing an incremental loss of 8010 km² over two decades. Future conflict was projected using the PLUS model, calibrated on 2010–2020 observed expansion and validated against the 2020 classified map (OA = 93.99%, Kappa = 0.91). Under the Business-as-Usual (BAU) scenario, 33,368 km² of currently open PV-suitable land faces urban encroachment by 2030; the Ecological Conservation Priority (ECP) scenario reduces this figure to approximately 30,767 km², while the Economic Development (ED) scenario yields a near-identical outcome to BAU, indicating that development velocity alone does not determine the spatial extent of conflict—the allocation of growth does. These findings provide a quantitative basis for designating energy-strategic reserve zones within national spatial planning frameworks and demonstrate that targeted spatial governance, applied at high-pressure locations, can substantially slow the erosion of the region’s solar energy land base. Full article

This study introduces two compact wideband substrate-integrated waveguide (SIW) filters fabricated using thin-film technology. The wideband bandpass response is achieved by incorporating interdigital capacitor (IDC) structures into a half-mode SIW (HMSIW) transmission line. An equivalent LC circuit model is formulated to analyze the influence of IDC parameters on the generation of transmission zeros. For the first filter (BPF 1), a third-order IDC coupling configuration is employed, resulting in a 1 dB passband spanning 11 GHz to 18 GHz, a minimum insertion loss of 0.66 dB, three transmission zeros that enhance stopband performance, and a compact core dimension of $0.49{\lambda }_g\times 0.29{\lambda }_g$. For further miniaturization, a modified HMSIW transmission line incorporating a metal-insulator-metal (MIM) capacitor at the equivalent magnetic wall is proposed. This design effectively reduces the transverse dimension of the waveguide while maintaining the original cutoff frequency. Utilizing this configuration, the second bandpass filter (BPF 2) was designed and fabricated employing double-layer ceramic thin-film technology. The resulting filter exhibits a 1 dB passband spanning 10 GHz to 18 GHz, a compact footprint measuring $0.44{\lambda }_g\times 0.23{\lambda }_g$, a minimum insertion loss of 0.58 dB, and features three transmission zeros. The fabricated and measured results of both filters show good agreement with simulations. Compared with previously reported wideband SIW filters, the proposed designs demonstrate comprehensive advantages in fractional bandwidth, insertion loss, out-of-band suppression, and circuit size, providing effective filtering solutions for high-density integration of microwave and millimeter-wave RF systems. Full article

Accurate prediction of compost maturity is vital for ensuring quality, safety, minimum substrate weight loss and agronomic performance of compost products. In this study, eight supervised machine learning (ML) classification models including Random Forest, Logistic Regression, Decision Tree, Gaussian and Multinomial Naive Bayes, K-Nearest Neighbors, Support Vector Machine, and AdaBoost were systematically evaluated for their ability to predict compost maturity using three key indicators: cation exchange capacity (CEC), carbon to nitrogen ratio (C/N), and humic acid (HA) content. A dataset comprising 756 samples (4 composting/vermicomposting systems × 7 treatments × 9 time points × 3 replicates) was generated. To reduce replicate-induced variability and ensure robust machine learning analysis, triplicates were averaged at each time point, resulting in 252 effective observations used for model development. Pearson correlation and heatmap analysis indicated strong interdependencies among CEC, HA, total nitrogen (TN) and organic matter (OM) content, confirming their collective utility in compost maturity classification. Model performance was assessed based on classification metrics (accuracy, precision, recall, F₁-score) and regression-based error indicators, including mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and coefficient of determination (R²). Ensemble models, particularly RF and AdaBoost, showed the highest predictive accuracy (up to 0.98) and lowest error rates (e.g., MAE < 0.05, RMSE < 0.1, R² > 0.95) when predicting CEC and C/N-based maturity classes. HA-based predictions showed slightly lower precision and higher variance across models. Full article