Browse Articles

Longitudinal study of Orthohantavirus hantanense in Apodemus agrarius and disease risk assessment in the Republic of Korea during 2000–2019

Progressive decomposition of infrared and visible image fusion network with joint transformer and Resnet

Abstract The objective of image fusion is to synthesize information from multiple source images into a single, high-quality composite that is information-rich, thereby enhancing both human visual interpretation and machine perception capabilities. This process also establishes a robust foundation for downstream image-related tasks. Nevertheless, current deep learning-based networks frequently neglect the distinctive features inherent in source images, presenting challenges in effectively balancing the interplay between basic and detailed features. To tackle this limitation, we introduce a progressive decomposition network that integrates Lite Transformer (LT) and ResNet architecture for infrared and visible image fusion (IVIF). Our methodology unfolds in three principal stages: Initially, a foundational convolutional neural network (CNN) is deployed to extract coarse-scale features from the source images. Subsequently, the LT is employed to bifurcate these coarse features into basic and detailed feature components. In the second phase, to augment the detail information across various inter-layer extractions, we substitute the conventional ResNet preprocessing with a combination of coarse and LT module. Cascade LT operations are implemented following the initial two ResNet blocks (ResB), enabling two-branch feature extraction from these reconfigured blocks. The final stage involves the design of specialized fusion sub-networks to process the basic and detail information blocks extracted from different layers. These processed image feature blocks are then channeled through semantic injection module (SIM) and Transformer decoders to generate the fused image. Complementing this architecture, we have developed a semantic information extraction module that aligns with the progressive inter-layer detail extraction framework. The LT module is strategically embedded within the ResNet network architecture to optimize the extraction of both basic and detailed features across diverse layers. Moreover, we introduce a novel correlation loss function that operates on the basic and detail information between layers, facilitating the correlation of basic features while maintaining the independence of detail features across layers. Through comprehensive qualitative and quantitative analyses conducted on multiple infrared-visible datasets, we demonstrate the superior potential of our proposed network for advanced visual tasks. Our network exhibits remarkable performance in detail extraction, significantly outperforming existing deep learning methodologies in this domain.

Transport equity assessment based on accessibility disparities in terms of multi-job opportunities across Beijing

Outcome prediction abilities of basketball players shooting free throws

Skilled athletes, such as basketball players, seem to anticipate the outcome of their actions, likely due to well-developed internal models that enhance predictive accuracy. This study explores whether experienced basketball players can verbally predict their free throw outcomes above chance, examining the role of expertise and potential biases in prediction accuracy. Therefore, 19 experienced basketball players performed 500 free throws in a laboratory setting. Within 2.5 seconds of visual and acoustic occlusion after ball release, they had to predict the result of every second trial as either a hit or a miss. Individual prediction accuracies above chance were calculated, and the hit bias was quantified by a signal detection theory measure (criterion c). Speech characteristics (latency and amplitude) were used as an additional proxy to investigate the prediction process. It was hypothesized that experienced players would make faster predictions of successful shots and would articulate these with greater volume compared to unsuccessful ones, reflecting the processing of available information about their throw execution and heightened response bias toward success. Results showed that participants were able to significantly predict their free throw outcomes above chance level. An earlier described hit bias could be replicated and was further observed as a louder and faster articulation of hits than misses. Overall, natural motor experts, in this case basketball players, seem to access their internal models and use the gathered information to predict the outcome of their free throws. However, they show a bias to predict hits, which is also manifested in the speech characteristics of latency and amplitude.

“In silico analysis of human TLR3 missense single nucleotide polymorphisms and their potential association with cancer”

Abstract Cervical cancer is a prevalent malignancy worldwide and represents a significant health burden for women. Toll-like receptors (TLRs) are crucial components of the innate immune system and play a vital role in recognizing pathogen-associated molecular patterns. Among the TLRs, TLR3 has been implicated in the defense against viral infections, including human papillomavirus (HPV), the primary etiological agent of cervical cancer. Missense single-nucleotide polymorphisms (SNPs) in the TLR3 gene can alter its protein structure and potentially influence its function, leading to variations in immune responses and disease susceptibility. This study aimed to investigate the impact of TLR3 missense SNPs on cervical cancer development through in silico methods. A comprehensive analysis was performed using various computational tools and databases to predict the functional consequences of identified TLR3 missense SNPs. The selection criteria for SNPs included their reported association with cervical cancer or their potential effects on TLR3 structure and function. Three extremely detrimental TLR-3 nsSNPs, namely N284I (rs5743316), C37R(rs752889035), and L360P(rs768091235), have been found among the 150 nsSNPs that have so far been reported in the dbSNP database. The in silico analysis suggests that these genetic variations may contribute to the development and progression of cervical cancer by modulating TLR3 function. Further experimental studies are warranted to validate these findings and elucidate the underlying mechanisms, which may aid in developing novel therapeutic strategies for cervical cancer prevention and treatment.

Effects of boysenberry on postprandial energy metabolism in healthy adults: A randomized controlled crossover trial

Brown adipose tissue (BAT) is essential for thermoregulation and energy metabolism, converting fatty acids into thermal energy in response to cold exposure and dietary intake, thereby contributing to both cold-induced thermogenesis and diet-induced thermogenesis (DIT). Our previous research suggests that boysenberry anthocyanins (BoyACs) may activate BAT under cold conditions, and we hypothesized that BoyACs could also influence DIT through the activation of BAT. This pilot randomized, double-blind crossover trial aimed to evaluate the effects of daily intake of BoyACs on DIT in healthy adults (registration number: UMIN000047413). Twenty-two participants consumed either a boysenberry juice (BoyJ) beverage containing 61.0 mg of BoyACs or a placebo beverage daily for four weeks, with a four-week washout period separating the two interventions. Three participants withdrew during the trial, resulting in data from 19 participants being analyzed. Results showed no significant changes in DIT, defined as increase in postprandial energy expenditure, or skin temperature of BAT regions. However, a significant increase in postprandial fat oxidation was observed. No significant differences were observed in other outcomes. These results suggest that BoyJ intake does not significantly affect postprandial energy expenditure but may influence substrate utilization to promote fat oxidation. Further studies focusing on substrate utilization, particularly fat oxidation, as the primary outcome are necessary to confirm these results and fully understand the implications of BoyJ intake on energy metabolism.

Reconfigurable control of coherence, dissipation, and nonreciprocity in cavity magnonics

Abstract Precise control of coupling strength, damping rate and nonreciprocity in photon–magnon systems is essential for advancing hybrid quantum technologies, including reconfigurable microwave components and quantum transducers. Here, we demonstrate magnetic field angle-dependent control of photon–magnon coupling and magnon dissipation in a cross-shaped microwave cavity supporting a spatially nonuniform radio-frequency (rf) magnetic field. By rotating the external magnetic field angle $$\theta$$ relative to the normal of the transmission line within the cavity plane, we simultaneously control the coherent coupling strength $${g}_{0}$$ , the ferromagnetic resonance (FMR) damping rate, and the system’s nonreciprocal response. The nonuniform rf field selectively excites both the uniform FMR mode and finite-wavevector spin waves in an Yttrium Iron Garnet (YIG) film, enabling angle-dependent two-magnon scattering. While typically regarded as a passive loss mechanism, we show that two-magnon scattering can serve as a dynamic and reversible knob to control magnon damping. Moreover, we realise nonreciprocity originating from the spatial asymmetry of the rf field, in addition to conventional phase-based schemes. These findings introduce new opportunities for in situ control of coherence, dissipation, and nonreciprocity in cavity magnonics, with broad implications for reconfigurable quantum and spintronic systems.

Raman spectroscopy identified fingernail compositional differences between sexes and age-related changes but not handedness or fingers in a healthy cohort

Background Nail properties and appearances can indicate a person’s underlying systemic diseases. Raman spectroscopy is an established laboratory technique and has been applied to nails, identifying spectral differences between healthy individuals and patient populations. Objective We aim to explore the importance of potential spectral or chemical variations in nails between sexes, age groups, hands, and fingers. Methods Twenty male and twenty female participants without known musculoskeletal or dermatological diseases donated nail clippings from each finger. The clippings were cleaned, and Raman spectra collected and analysed using a standardised protocol. Results In total 2000 spectra were collected. Females have higher intensities of disulphide, protein, and lipid bands, particularly in their 40s, than males. Age-related changes were prominent in female nails, especially in sulphur-related bands. No significant differences were observed between nails from the left and right hands or among different fingers. Limitations We did not control other factors such as diet, medication, or different occupation or sports participation. Conclusion This is the first study to use Raman spectroscopy to compare nail composition across different ages and sexes in healthy adults. The findings provide a strong basis for further studies on nails at the population level for screening or monitoring diseases.

Leveraging internet use for sustainable agriculture: the impact of digital training on adoption of energy-smart agricultural practices and welfare

An unclean slate, discrepancies between food input and recovered protein signal from experimental foodcrusts

Organic residues are a rich source of biomolecular information on ancient diets. In particular, foodcrusts, charred residues on ceramics, are commonly analysed for their lipid content and to a lesser extent protein in order to identify foods, culinary practices and material culture use in past populations. However, the composition of foodcrusts and the factors behind their formation are not well understood. Here we analyse proteomic data (available via ProteomeXchange with identifier PXD059930) from foodcrusts made using a series of mixtures of protein- (salmon flesh), lipid- (beef fat) and carbohydrate-rich (beetroot) foods to investigate the relationship between the biomolecular composition of the input and the recovered signal using conventional methods applied to archaeological material. Additionally, using 3D modelling we quantify the volume of foodcrust generated by different ingredient combinations The results highlight biases in the data obtained in the analyses of organic residues both in terms of identified resources reflecting the cooked foodstuffs, e.g., an overrepresentation of fish proteins, as well as with regards to the abundance of foodcrust, for example mixtures of only salmon and beef fat resulted in relatively small amounts of foodcrust, and suggest caution in interpreting the composition of residues formed from complex mixtures of foodstuffs.

Medical triage as an AI ethics benchmark

Abstract We present the TRIAGE benchmark, a novel machine ethics benchmark designed to evaluate the ethical decision-making abilities of large language models (LLMs) in mass casualty scenarios. TRIAGE uses medical dilemmas created by healthcare professionals to evaluate the ethical decision-making of AI systems in real-world, high-stakes scenarios. We evaluated six major LLMs on TRIAGE, examining how different ethical and adversarial prompts influence model behavior. Our results show that most models consistently outperformed random guessing, with open source models making more serious ethical errors than proprietary models. Providing guiding ethical principles to LLMs degraded performance on TRIAGE, which stand in contrast to results from other machine ethics benchmarks where explicating ethical principles improved results. Adversarial prompts significantly decreased accuracy. By demonstrating the influence of context and ethical framing on the performance of LLMs, we provide critical insights into the current capabilities and limitations of AI in high-stakes ethical decision making in medicine.

DSTF-GKAN: A lightweight spatiotemporal fusion framework for real-time eavesdropping detection in dynamic smart grid networks

With the rapid development of smart grids and the Power Internet of Things (PIoT), wireless communication networks are facing the severe threat of dynamic eavesdropping attacks. Traditional detection methods rely on static assumptions or shallow models, which are not capable of dealing with complex topology mutations and high-dimensional nonlinear features. There is an urgent need for efficient and lightweight adaptive solutions. This study proposes a Dynamic Spatiotemporal Fusion Framework (DSTF-GKAN), which integrates the spatiotemporal dynamic modeling capability of Graph Recurrent Neural Networks (GRNN) with the lightweight adaptive spline approximation mechanism of Kolmogorov-Arnold Networks (KAN). By adaptively optimizing the mesh to dynamically adjust the spline control points and introducing hierarchical sparse regularization to compress parameters, the model enhances its sensitivity to channel anomalies through the integration of physical layer security (PLS) feature constraints. Experimental results show that under dynamic scenarios with an attack mutation rate (AMR = 0.5), DSTF-GKAN achieves a detection F1 score of 0.891, which is a 7.1% improvement over GRNN, and reduces the localization error (RMSE = 0.518 m) by 16.2%. After quantization and pruning optimization, the model has a parameter size of only 0.2 MB, with an inference latency of 0.9 ms and energy consumption of 16mJ on edge devices. Ablation experiments have verified the necessity of the GRU-GCN module (contributing 4.9% to the F1 score) and PLS regularization (improving the F1 score by 1.3%). DSTF-GKAN provides an efficient, robust, and interpretable detection framework for smart grid security. Its lightweight design promotes real-time edge defense and lays the theoretical and technical foundation for the construction of a secure energy internet ecosystem.

Impact of solid particle geometry, size, and intensity, coupled with fluid velocity, on erosion dynamics in elbow conduits

Assessing antibiotics consumption, use and outcomes in a Yemeni tertiary hospital: A prospective cross-sectional study

Background Antibiotics (ABs) have saved countless lives, but their misuse has led to a serious problem: antibiotic resistance. This growing phenomenon poses serious threats to public health worldwide, as it could make treating infections significantly more difficult in the future. Objectives This study aimed to investigate antibiotic consumption and use patterns in a tertiary hospital in Sana’a, Yemen, by comparing Prescribed Daily Doses (PDD) to Defined Daily Doses (DDD), and identifying factors associated with antibiotic misuse and its impact on patient outcomes. Methods A prospective cross-sectional study was conducted among adult inpatients in a tertiary hospital in Sana’a, Yemen over two months (January 12 to March 11, 2024), involving 597 patients. Data on antibiotic prescriptions, patient demographics, and outcomes were collected. Results A high prevalence of antibiotic use was observed (92.5%), with a notable proportion of prescriptions from the “Watch” category (56.7%). Significant PDD-DDD deviations were common, encompassing both overuse (36.8%) and underuse (63.2%). Factors associated with antibiotic deviations included patient age (26–44 years), gender (female), and ward type (private). The most commonly prescribed antibiotics were Ceftriaxone (33.6%), Metronidazole (21.8%), Vancomycin (6.0%), Levofloxacin (4.9%), Imipenem/Cilastatin (4.7%), and Moxifloxacin (3.6%). Notable deviations from DDD were observed for Levofloxacin (overuse by 28%), Imipenem/Cilastatin (underuse by 40.5%), and other agents. Antibiotic misuse was associated with longer hospital stays and less favorable discharge outcomes. Conclusion The study found an alarmingly high prevalence of antibiotic use and excessive consumption, with both overuse and underuse patterns observed, underscoring the need for effective regulatory interventions and improved antibiotic stewardship in Yemen.

The impact of the BDNF Val66Met genotype on intrusive memories following trauma exposure and in PTSD is moderated by sex and timing of trauma exposure

Abstract Intrusive memories are a key symptom of Post-Traumatic Stress Disorder (PTSD). Brain Derived Neurotrophic Factor (BDNF) has been proposed as a possible mechanism influencing intrusive memories in PTSD given its role in synaptic plasticity and memory consolidation. The BDNF Val66Met polymorphism has been linked PTSD susceptibility and episodic memory disturbances however previous research outcomes have been variable, potentially due to a failure to control for important confounds such as sex, ethnicity, BMI, developmental stage and extent of previous trauma experiences. This study explored the relationship between the BDNF Val66Met genotype and emotional memory (intrusive memories and recall) in PTSD controlling for these factors in 276 participants: 53 with PTSD, 118 Trauma Exposed and 105 Controls. Key findings revealed the PTSD group experienced significantly more negative intrusions than Controls, and females more intrusions than males, however there were no group or sex differences in negative memory recall. When developmental stage of trauma was considered in a traumatised sub-sample, BDNF genotype significantly interacted with PTSD status, sex, and developmental trauma stage. This highlights the importance of controlling for sex and timing of trauma on BDNF expression in neurobiological PTSD research, however further research is needed to replicate these preliminary findings and investigate the specific epigenetic and neurobiological mechanisms involved.

Investigation of wear behaviour and surface analysis of a coated H13 material for friction drilling application

In recent years, industries have seen many advancements in finding proper tools for machining to enhance productivity. Choosing a proper friction drilling tool that minimizes surface damage and improves tool life and productivity is essential. In this study, the wear characteristics of H13 steel among four samples (untreated, heated, TiAlN, and AlCrN) were investigated through a pin-on-disc machine, focusing on highlighting the wear behaviour and surface morphology. The novelty of this study is to analyze an optimal friction drilling tool that can enhance its life. The tempering process was carried out to improve the hardness of the H13 steel tool from 37 HRC to 57 HRC. During the wear test process, the temperature is maintained at 250°C. Using an Atomic Force Microscope (AFM), the worn surface of the samples was analyzed. Among the four samples (untreated, heated, TiAlN, and AlCrN), the untreated samples were affected by adhesive wear and oxidation. It is observed that the tempering helps the coated H13 samples to appear wear-resistant; the material loss obtained for the coated samples is much less compared to the uncoated samples. The untreated and heated sample CoF values observed are 0.713 and 0.591; for TiAlN and AlCrN, the CoF values observed are 0.481 and 0.416. This study reveals that AlCrN Coated H13 steel exhibited the best wear response. Hence, it is suitable for Friction drilling applications.

Spatio-temporal dynamics of ingroup interactions in macaques

Abstract When sharing a space with others, many species including humans evolved a compromise regulating occupancy influenced by social determinants. For example, students in a classroom tend to sit close to their friends, keeping the same spots across days, revealing the social structure in the classroom. This place preference suggests that factors such as social hierarchy and affiliation can shape space utilization, contrasting with random walk models of agents moving at random in any given direction. Here, we asked whether spatial occupancy of macaques within two unisex groups of four individuals, reveals a structured space utilization beyond simple spatial affordance within the finite space. To this end, in two groups of four animals, we analyzed the simultaneously recorded positions of each individual while the group roamed in an enclosure. The data was gathered using automated devices that allow measuring accurate simultaneous positions and calculating precise inter-individual distance, which is impossible in classical ethology, even using GPS devices. Thus, our setup opens new possibilities for modelling approaches, to characterize social interaction dynamics in small enclosures. We found that (1) The identity of each animal could be decoded from its individual pattern of spatial occupancy, revealing that each animal sustained a consistent spatial footprint across multiple days. (2) The average distance between monkeys was a proxy for their social hierarchy, confirming that interpersonal distance is correlated with affiliation and dominance hierarchy. (3) Alternating the social context by removing one of the monkeys revealed that only removing the closest social partner influenced occupancy. (4) Finally, the distribution of distance between pairs of monkeys was bimodal and was modeled using a random walk approach with an additional parameter reflecting the propensity to stay in close proximity, which was again related to dominance hierarchy. These analyses reveal that space utilization is structured as a function of social determinants in macaques and demonstrate the usefulness simple modeling approaches to further study group organization in neuro-ethological settings.

Shifting respiratory pathogens: Post-COVID-19 trends in community-acquired infections in underserved communities

Respiratory tract infections, caused by various bacteria and viruses, pose a significant global health burden. In Lebanon, post-COVID-19 epidemiological data on respiratory infections remain scarce. To address this gap, this multicenter study investigates the epidemiology of community-acquired acute respiratory infections among children and adults in Tripoli, North Lebanon. From May 2023 to February 2024, nasopharyngeal samples were collected from outpatients with acute respiratory infections visiting hospitals and pediatric clinics in Tripoli. Samples were analyzed using BioFire® Respiratory Panel 2.1 Plus (bioMérieux, France), which targets 23 pathogens, including 19 viruses and four bacteria. We used multivariable logistic regression models to identify the determinants of respiratory infections and examine associations between respiratory pathogens. Among 324 enrolled patients, 69.1% were co-infected with at least one pathogen. Human rhinovirus/enterovirus was the most prevalent (27.2%), followed by influenza A (19.8%), particularly influenza A/H1-2009 (16.4%), and RSV (11.4%). SARS-CoV-2 was still circulating with a prevalence of 6.8%. Classical human coronaviruses accounted for 6.1% of infections, with HCoV-NL63 (2.8%) being the most common. Parainfluenza viruses were identified in 5.2% of patients, with type 4 (2.5%) being the most prevalent, followed by type 3 (1.5%), type 1 (1.2%), and type 2 (0.3%). Logistic regression analysis revealed that human rhinovirus/enterovirus infection decreased the likelihood of influenza A (OR=0.25; 95%CI = 0.10–0.54; P = 0.001) or SARS-CoV-2 (OR=0.21; 95%CI = 0.03–0.75; P = 0.039) co-infection. Additionally, our logistic regression models identified significant associations between various determinants, symptoms, and common viruses, including a lower likelihood of influenza A (OR=0.23; 95%CI = 0.06–0.76; P = 0.019) and RSV (OR=0.29; 95%CI = 0.10–0.76; P = 0.017) infection among patients with higher educational levels. Notably, parainfluenza virus infections occurred significantly more in refugee patients (OR=7.22; 95%CI = 1.19–37.0; P = 0.020) compared to the host community. In conclusion, this study provides critical insights into the post-pandemic epidemiology of respiratory infections in Lebanon, informing clinicians, health authorities, and policymakers to optimize diagnostics, preventive measures, and antimicrobial stewardship strategies.

Spatio-temporal patterns in growing bacterial suspensions

Corneal biomechanical predictors of intraocular pressure elevation after intravitreal anti-VEGF injection

Purpose To investigate whether corneal biomechanical parameters measured via Corvis ST can predict acute intraocular pressure (IOP) elevation following intravitreal anti-VEGF injection. Design Retrospective observational study. Subjects Forty eyes from patients with neovascular age-related macular degeneration or retinal vein occlusion who underwent anti-VEGF therapy. Methods IOP was measured using the Corvis ST immediately before and 10 minutes after injection. The following biomechanical parameters were evaluated: DA Ratio MAX (2mm), biomechanically corrected IOP (bIOP), Peak Distance, Deflection Amplitude Max, Integrated Radius, and Stress-Strain Index (SSI). Main outcome measures Acute post-injection IOP elevation (continuous) and IOP spikes ≥10 mmHg (binary). Results The mean IOP increased significantly from 14.5 ± 3.17 to 24.7 ± 7.44 mmHg post-injection (p < 0.0001). IOP spikes ≥10 mmHg occurred in 55% of eyes. On multivariate analysis, higher bIOP (β = +1.17, p = 0.048) and lower DA Ratio MAX (β = –5.40, p = 0.038) were independent predictors of IOP elevation. DA Ratio MAX was the only significant predictor of IOP spikes (OR = 0.70, 95% CI: 0.51–0.96, p = 0.035). ROC analysis showed that DA Ratio MAX alone (AUC = 0.739) outperformed bIOP (AUC = 0.607), with the combined model yielding the highest AUC (0.773). A cutoff of DA Ratio MAX ≤4.936 provided 81.8% sensitivity and 42.9% specificity for predicting spikes. Conclusions DA Ratio MAX (2mm), reflecting global ocular compliance, was a significant predictor of acute IOP spikes after anti-VEGF injection. Alongside bIOP, it may be useful for pre-injection risk stratification of pressure-related complications.