Browse Articles
Discover research articles across all indexed journals
Association between smartphone overdependence and alcohol and tobacco use behaviors among adolescents in Korea
Red light therapy: the science behind the hype
Improved ecological risk assessment of phenol in sediments via species sensitivity distribution and equilibrium partitioning method using water toxicity data
AI ‘scientists’ promise to accelerate research — how do they work?
A novel ANN-based approach for fault detection and classification in modern TCSC-compensated transmission lines integrated with DFIG-based wind farms utilizing WST
Abstract The dynamic characteristics of doubly fed induction generator (DFIG)-based wind farms, together with the variable reactance introduced by a thyristor-controlled series capacitor (TCSC) compensation, significantly alter fault current profiles and apparent line impedance, which may lead to maloperation of conventional transmission line protection schemes. In this paper, an intelligent fault detection and classification approach for a TCSC-compensated transmission line integrated with a DFIG-based wind power generation system is proposed. The method used the Wavelet Scattering Transform (WST) for robust feature extraction with a feed-forward back propagation neural network (BPNN) for accurate classification. The proposed approach exploits the inherent stability and invariance properties of WST to extract discriminative features from transmission line current signals under dynamic operating conditions. Its performance is evaluated through extensive simulation studies involving 3240 fault cases covering variations in fault type, location, inception angle, wind-farm operating conditions, and TCSC compensation levels. A comparative analysis with discrete wavelet transform (DWT)-based features is conducted in terms of detection latency, robustness, computational efficiency, and classification accuracy. Simulation results demonstrate that the WST-based approach outperforms conventional DWT-based methods, achieving high classification accuracy up to 100% with strong robustness to noise and operating variations, while maintaining a response time suitable for practical protection applications. These results confirm the effectiveness of the proposed scheme for modern series-compensated transmission systems with high renewable energy penetration. All simulations are carried out using MATLAB.
Bioinformatics and experimental validation of druggable targets in non-alcoholic fatty liver disease
Effect of tipiracil hydrochloride, thymidine phosphorylase inhibitor, on the ischemia/reperfusion injury of brain tissue in rats
Abstract Thymidine phosphorylase (TP) expression is increased in neurons under ischemia/reperfusion (I/R) conditions. Our aim was to evaluate the effect of tipiracil hydrochloride (TPI), a selective TP inhibitor, on rat brain tissue subjected to I/R. Both common carotid arteries were occluded for 30 min in the ischemic untreated group of rats (C-IR), and ischemic groups treated with tipiracil 25 mg/g (T-IR25) or 50 mg/kg (T-IR50). In the control group (C), the arteries were not ligated. Tipiracil was given during ischemia, and after 8 h of I/R intraperitoneally. After 24 h of I/R, brain tissue was isolated for histology and immunohistochemy of TP expression. Metalloproteinases 2 and 9 (MMP-2 and -9) and tissue inhibitor of metalloproteinases (TIMP-1) were determined in serum at 3 and 24 h of reperfusion. TP expression in brain tissue was the highest in C-IR and T-IR25 compared to the C and T-IR50 . No changes in serum TP levels were observed. After 24 h, there was a significant decrease in MMP-9 levels in T-IR25 compared to the C-IR and T-IR50. MMP-2 levels also decreased significantly at this time point in all groups compared to group C, which correlated with increased TIMP-1 activity in the T-IR25 and T-IR50. The inhibition of TP activity in the group receiving TPI suggests its protective effect on brain tissue under I/R conditions. The decrease in MMP activities in the treated groups suggests a protective effect of TPI on the development of neuroinflammation caused by local brain tissue ischemia.
Tumor implantation site dictates the immune landscape of syngeneic mouse models of head and neck cancer
Feature-specific threat coding in lateral septum guides defensive action
Genome editing-based refinement of GPCR visualization in mice using the oxytocin receptor as a model
Tear biomarker changes and ocular surface recovery with low-level light therapy after cataract surgery: a double-masked randomized controlled clinical trial
Abstract Dry eye disease (DED) is a common complication following cataract-surgery, potentially impairing visual recovery. Low-level light therapy (LLLT) has emerged as a noninvasive approach to improve ocular surface status. This study evaluated the perioperative effects of LLLT on ocular surface modulation by correlating clinical outcomes with tear biomarker dynamics and developing predictive models to identify patients most likely to benefit from LLLT. Of the 98 patients initially randomized, 88 were included in the final analysis, with 44 allocated to the LLLT group and 44 to the sham treatment group. Clinical evaluation—including Ocular Surface Disease Index (OSDI) as the primary outcome, and tear breakup time, Schirmer test, and tear osmolarity as secondary endpoints—was performed preoperatively and one month postoperatively to classify patients as preclinical or having DED. Tear levels of biomarkers involved in tissue repair and neurotrophic-signaling (GDF-15,β-NGF, VEGFA, PDGF-AB, PDGF-CC) and inflammation (OPN, OPG, TNF-α) were assessed as exploratory endpoints and quantified using Luminex-FlexMap3D technology. LLLT-treated patients showed significantly higher clinical improvement post-cataract surgery than sham controls (44.2% vs. 4.4%; p < 0.0001). In DED patients, LLLT significantly increased GDF-15 (77.6 ± 7.2 to 95.5 ± 7.4 pg/mL; p = 0.0112) and PDGF-CC (711.4 ± 76.6 to 1024 ± 130.8 pg/mL; p = 0.0473). The LLLT-induced β-NGF increase was more pronounced in preclinical cases than in those with baseline DED (10.1 ± 3.02 vs. 5.8 ± 0.55 pg/mL; p = 0.0271). Biomarker profiles indicated inflammatory resolution post-LLLT, whereas sham-treated eyes showed persistent inflammation. Finally, integrative modeling of clinical and molecular data yielded a LASSO-adjusted AUC of 0.886, underscoring high discriminative performance. LLLT accelerates ocular surface recovery after cataract surgery by modulating reparative, neurotrophic, and inflammatory pathways. Tear biomarkers assessed at baseline and through one-month dynamics (GDF-15, PDGF-CC, β-NGF) and integrated with clinical parameters, may help predict treatment response and guide personalized postoperative management.
Machine learning based detection of covert communications under jamming interference
Abstract In large-scale Internet of Things (IoT) networks, detecting covert communications—hidden transmissions that evade monitoring—is essential to prevent misuse of wireless infrastructure. However, challenges such as fading, noise, jamming interference, and unpredictable traffic complicate reliable detection by a monitoring node (Willie). This paper introduces a hybrid analytical-machine learning (ML) framework for robust detection of covert signals under jamming and Rayleigh fading conditions. An analytical energy detection model is first derived to compute false alarm and missed detection probabilities, establishing a baseline and informing Monte Carlo simulations for dataset generation. The real and imaginary components of simulated complex baseband signals are extracted as features, enabling supervised training of Decision Tree (DT) and Random Forest (RF) classifiers without prior knowledge of transmit powers. Evaluations demonstrate that both ML models outperform the analytical benchmark, with RF achieving a 26.8% reduction in total detection error at a distance of d = 1 km. Further assessments across varying transmitter and jammer power levels confirm the framework’s robustness in interference-limited environments. By integrating theoretical modeling for data credibility with data-driven ML for adaptive classification, this approach provides a scalable, power-agnostic solution for securing real-world wireless networks against covert threats.
Establishment of an immortalized erythroid model for Hb Bart’s hydrops fetalis with homozygous α0-thalassemia Southeast Asian deletion
Gene expression profiling by RNA-sequencing reveals regulators of intramuscular fat in Black Slavonian pigs
Abstract Intramuscular fat (IMF) plays an important role in determining meat quality traits such as flavor, tenderness, and juiciness. While numerous studies have investigated the genetic basis of IMF in commercial pig breeds, data on local breeds remain rather limited. In this study, we used RNA-sequencing to characterize the transcriptomic differences between high-IMF and low-IMF Black Slavonian pigs, a native Croatian breed known for superior meat quality. Muscle samples ( Longissimus thoracis et lumborum ) from 14 pigs with divergent IMF levels were collected shortly after slaughter, preserved in liquid nitrogen, and stored at − 80 °C until RNA extraction. Intramuscular fat content was determined from the same muscle 24 h post mortem using the Soxhlet extraction method (ISO 1443:1973). These samples were then analyzed to identify differentially expressed genes (DEGs) and enriched pathways. A total of 519 genes were differentially expressed ( p ≤ 0.05), with 457 remaining significant after false discovery rate correction. The high-IMF group exhibited upregulation of genes associated with lipid metabolism (e.g., SCD, ADIPOQ, CIDEC, FABP4), PPAR signaling, and adipogenesis, while genes linked to muscle structure and oxidative metabolism were downregulated. Functional enrichment and gene set enrichment analyses highlighted coordinated regulation of pathways related to fatty acid biosynthesis, extracellular matrix (ECM) remodelling, angiogenesis, and Notch signaling. Notably, several ECM-related genes (LAMA1, TIMP4) and angiogenic factors (FGF2, NRP1) were significantly upregulated, suggesting that adipocyte expansion in muscle requires parallel vascular and structural adaptations. Importantly, several of the top 30 DEGs, including EHD2, NRG4, UTRN, FLNA, and HMCN1, represent novel candidate genes not previously linked to IMF in pigs, pointing to potential breed-specific mechanisms. These genes are associated with membrane trafficking, paracrine signalling, cytoskeletal re-modelling, and ECM dynamics. Our findings contribute new molecular insights into IMF regulation in local pig breeds and provide a foundation for developing targeted breeding strategies to improve pork quality through intramuscular fat enhancement.
Evaluating patients’ trust in health information based on different dimensions of trust
SHAP analysis of an improved EEG-based mental workload classification framework: utilizing data augmentation and explainable AI
Abstract Mental workload (MWL) classification using electroencephalogram (EEG) signals is crucial for cognitive neuroscience and is also a challenging research area in brain-computer interface (BCI). Since the EEG signals fluctuate a lot across sessions and individuals, there is a need for a robust classification model that generalizes well for real-world applications. In this work, we used the publicly available dataset “An EEG dataset for cross-session mental workload estimation: passive BCI competition of the Neuroergonomics Conference 2021”, and the standard EEGNet model to classify the MWL into three classes (Low, Med, and High). To improve the performance of the model, a synthetic minority oversampling technique (SMOTE) was used by creating synthetic EEG samples, and key hyperparameters ( F 1 , F 2 , and D ) of EEGNet were systematically varied to identify the optimal configuration. Furthermore, Shapley Additive Explanations (SHAP) analysis was performed to identify the most influential EEG channels for model prediction. The proposed approach achieves the highest accuracy of 80.5% and 82.7% without and with SMOTE, respectively. The comparative analysis showed that applying SMOTE resulted in an average performance improvement of approximately 3%. A Wilcoxon signed-rank test confirmed that this improvement was statistically significant ( p < 0.05). Finally, the SHAP analysis revealed that the most informative EEG channels were located over the parieto-occipital and temporal regions, which is consistent with established neurophysiological evidence related to MWL processing. The proposed framework improves both performance and explainability in EEG-based MWL classification, representing a systematic integration of SMOTE and SHAP analysis.