Search Results (12,369)

It has been demonstrated that interactive speech and noise modeling outperforms traditional speech modeling-only methods for speech enhancement (SE). With a dual-branch topology that simultaneously predicts target speech and noise signals and employs bidirectional information communication between the two branches, the quality of the enhanced speech is significantly improved. However, the dual-branch topology greatly increases the model complexity and deployment cost, thus limiting its practicality. In this paper, we propose UniInterNet, a unidirectional information interaction-based dual-branch network to achieve noise modeling-assisted SE without any increase in complexity. Specifically, the noise branch still receives information from the speech branch to achieve more accurate noise modeling. Subsequently, the noise modeling results are utilized to assist the learning of the speech branch during backpropagation, while the speech branch no longer receives the auxiliary information from the noise branch, so only the speech branch is required during model deployment. Experimental results demonstrate that under the causal inference condition, the performance of UniInterNet only marginally decreases compared to the corresponding bidirectional information interaction scheme, while the model inference complexity is reduced by about 75%. With comparable overall performance, UniInterNet also outperforms previous interactive speech and noise modeling-based benchmarks in terms of causal inference and model complexity. Furthermore, UniInterNet surpasses other existing competitive methods. Full article

Coffea canephora genotypes adopt distinct strategies to cope with drought and rehydration. We hypothesized that the greater drought tolerance of genotype ‘3V’ compared to ‘A1’, previously reflected in physiological and anatomical leaf traits after two water-stress (WS) cycles, could also be observed in P–V curve responses, root and branch anatomy, leaf midrib elongation (CVL), and root distribution. The ‘3V’ and ‘A1’ plants were grown under well-watered (WW) conditions and two cycles of water stress (WS). The ‘3V’ was more sensitive to WS, with reduced branch xylem vessel density (BXVD), while ‘A1’ demonstrated increased BXVD. Root xylem vessel area (RXVA) decreased to a greater extent in ‘3V’ than in ‘A1’, and both genotypes showed increased bulk elastic modulus. Regardless of water conditions, ‘A1’ maintained a higher relative leaf water content at the turgor loss point (RWC_TLP). Morphological acclimation did not occur in the second WS cycle. The ‘3V’ plants developed greater root mass in deeper soil layers than ‘A1’ under the WS condition. These findings suggest that ‘A1’ follows a conservative drought-avoidance strategy with lower physio-morphological plasticity, while ‘3V’ exhibits greater drought tolerance. Such responses highlighted coordinated physiological, morphological, and anatomical adaptations of the above- and below-ground organs for resource acquisition and conservation under WS. Full article

Bamboo is an efficient silicon accumulator with diverse phytolith morphotypes and composition. The bamboo culm sheath, traditionally considered as a modified leaf, plays a key role in bamboo taxonomy and provides significant mechanical and physiological support for shoot development, but its silicon deposition and phytolith morphological variation remain underexplored. We investigated silicon variation and phytolith morphology in D. brandisii culm sheaths at different growth stages. The results showed that silicon deposition in D. brandisii culm sheaths at different growth stages was comparable to foliage leaves but significantly greater than branches as in previous research. Phytolith concentration in the culm sheath blades of D. brandisii was higher, associated with their greater silicon content than the sheath bodies. Silicon precipitated and phytoliths were produced as the culm sheath matured. Silicon and phytolith contents were significantly greater in upper culm sheath bodies. D. brandisii culm sheaths were characterized by a high proportion of rondel phytoliths, whereas bilobate and bulliform flabellate phytoliths were not observed. Circular and saddle phytoliths accounted for a very low proportion. Stomata phytoliths were abundant in culm sheaths at the shooting stage and increased with sheath maturation, which supported the transpiration, respiration and photosynthesis in culm sheaths of the shoots. Elongate and acute phytoliths were extremely abundant in D. brandisii culm sheaths and increased with sheath maturation, which enhanced the mechanical and protective role of the culm. The phytolith morphotypes in D. brandisii culm sheaths did not demonstrate a specific trend with sheath position. Saddle phytoliths showed insignificant variation in D. brandisii culm sheaths. Full article

Recent experiments have reported distinct handedness of spin waves across the compensation temperatures of ferrimagnets, offering promising functionalities for ferrimagnet-based magnonic applications with two distinct polarizations. This paper investigates the effects of various factors on the compensation points of GdFe ferrimagnets through atomistic-level spin dynamics simulations. The results show that as the Gd composition increases, both the magnetization compensation temperature and the angular momentum compensation temperature of the GdFe alloy increase, with a linear relationship observed between the two compensation temperatures. Furthermore, we show that external magnetic fields and antiferromagnetic exchange strength can also modulate the compensation temperatures. Moreover, the antiferromagnetic exchange strength also affects the resonance frequency of ferrimagnetic materials. In the absence of an external field, the resonance frequency of GdFe is divided into two branches and both increase linearly with the increase in antiferromagnetic exchange strength. This study may stimulate fundamental research on compensated ferrimagnets, which may be useful for building chirality-based spintronics. Full article

Sugars will eventually be exported transporter (SWEET), a class of glucose transport proteins, is crucial in plants for glucose transport by redistribution of sugars and regulates growth, development, and stress tolerance. Although the SWEET family has been studied in many plants, little is known about its function in winter B. rapa (Brassica rapa L.). Bioinformatics approaches were adopted to identify the SWEET gene (BraSWEETs) family in B. rapa to investigate its role during overwintering. From the whole-genome data, 31 BraSWEET genes were identified. Gene expansion was realized by tandem and fragment duplication, and the 31 genes were classified into four branches by phylogenetic analysis. As indicated by exon–intron structure, cis-acting elements, MEME (Multiple EM for Motif Elicitation) motifs, and protein structure, BraSWEETs were evolutionarily conserved. According to the heat map, 23 BraSWEET genes were differentially expressed during overwintering, revealing their potential functions in response to low-temperature stress and involvement in the overwintering memory-formation mechanism. BraSWEET10 is mainly associated with plant reproductive growth and may be crucial in the formation of overwintering memory in B. rapa. The BraSWEET10 gene was cloned into B. rapa (Longyou-7, L7). The BraSWEET10 protein contained seven transmembrane structural domains. Real-time fluorescence quantitative PCR (qRT-PCR) showed that the BraSWEET10 gene responded to low-temperature stress. BraSWEET10 was localized to the cell membrane. The root length of overexpressing transgenic A. thaliana was significantly higher than that of wild-type (WT) A. thaliana under low temperatures. Our findings suggest that this gene may be important for the adaptation of winter B. rapa to low-temperature stress. Overall, the findings are expected to contribute to understanding the evolutionary links of the BraSWEET family and lay the foundation for future studies on the functional characteristics of BraSWEET genes. Full article

Background/Objectives: Alzheimer’s disease (AD) is the most common form of dementia, characterized by an irreversible decline in cognitive function. The pathogenesis of several neurodegenerative disorders has been linked to changes in the gut microbiota, transmitted through the gut-brain axis. Methods: We set out to establish by case-control study methodology whether there were any differences in the composition and/or function of the gut microbiota between older resident adults in care homes with or without an AD diagnosis via analysis of the microbial composition from fecal samples. Results: The microbial composition, determined by 16S rRNA gene profiling, indicated that AD sufferers had significantly increased proportions of Escherichia/Shigella and Clostridium_sensu_stricto_1, and significantly decreased proportions of Bacteroides, Faecalibacterium, Blautia, and Roseburia species. The increase in potentially pro-inflammatory bacteria was consistent with slightly higher concentrations of calprotectin, a biomarker of gut inflammation. Fecal concentrations of most microbial metabolites measured were similar across groups, although participants with AD had significantly increased proportions of the branched-chain fatty acid, iso-butyrate, and lower overall concentrations of total short chain fatty acids. Conclusions: Participants with Alzheimer’s disease have several key differences within their gut microbiota profile, in contrast to care home residents without Alzheimer’s disease. The altered microbiome included both compositional and functional changes linked to poorer health and gut inflammation. Full article

Flexible DC grids are an important technological means for optimizing power supply structures and promoting energy transition. However, as a system with low inertia and weak damping, the flexible DC grid inherently faces challenges, such as rapid rising of fault currents, vulnerability to significant damage, difficulty in fault interruption, and with regard to the poor overcurrent-withstanding capabilities of power electronic devices. To address these issues, this paper proposes a method for calculating the single-pole ground fault current in a symmetrical monopolar DC grid, and further introduces a matrix exponential calculation method. This method enables quantitative analysis of the influence of various component parameters on the fault current, taking into account the dynamic characteristics of both the faulted and healthy poles in the DC system. The results demonstrate the high accuracy of this calculation method. The analysis reveals that the inductance of the faulted branch has the greatest impact on the fault current, while the inductances of the adjacent outgoing lines also have a certain influence. In contrast, the inductances of lines not adjacent to the faulted branch have minimal impacts on the fault current. Furthermore, the grounding electrode parameters of the converter station connected to the faulted branch exert the most significant influence on the fault current, with the grounding electrode parameters of neighboring converter stations also showing a notable effect. This indicates that the fault current is impacted by the topology of the nearby DC grid, but is not affected by the fault currents at remote converter stations. Full article

In this paper, we investigated the role of generative AI in education in academic publications extracted from Web of Science (3506 records; 2019–2024). The proposed methodology included three main streams: (1) Monthly analysis trends; top-ranking research areas, keywords and universities; frequency of keywords over time; a keyword co-occurrence map; collaboration networks; and a Sankey diagram illustrating the relationship between AI-related terms, publication years and research areas; (2) Sentiment analysis using a custom list of words, VADER and TextBlob; (3) Topic modeling using Latent Dirichlet Allocation (LDA). Terms such as “artificial intelligence” and “generative artificial intelligence” were predominant, but they diverged and evolved over time. By 2024, AI applications had branched into specialized fields, including education and educational research, computer science, engineering, psychology, medical informatics, healthcare sciences, general medicine and surgery. The sentiment analysis reveals a growing optimism in academic publications regarding generative AI in education, with a steady increase in positive sentiment from 2023 to 2024, while maintaining a predominantly neutral tone. Five main topics were derived from AI applications in education, based on an analysis of the most relevant terms extracted by LDA: (1) Gen-AI’s impact in education and research; (2) ChatGPT as a tool for university students and teachers; (3) Large language models (LLMs) and prompting in computing education; (4) Applications of ChatGPT in patient education; (5) ChatGPT’s performance in medical examinations. The research identified several emerging topics: discipline-specific application of LLMs, multimodal gen-AI, personalized learning, AI as a peer or tutor and cross-cultural and multilingual tools aimed at developing culturally relevant educational content and supporting the teaching of lesser-known languages. Further, gamification with generative AI involves designing interactive storytelling and adaptive educational games to enhance engagement and hybrid human–AI classrooms explore co-teaching dynamics, teacher–student relationships and the impact on classroom authority. Full article

Stimuli-responsive polymeric nanostructures are compelling vectors for a wide range of application opportunities. The objective we sought was to broaden the array of self-assembling amphiphilic copolymers with stimuli-responsive characteristics by introducing a hydrophilic tunable monomer, (2-dimethylamino)ethyl methacrylate (DMAEMA), together with a hydrophilic one, lauryl methacrylate (LMA), within linear and branched copolymer topologies. Size exclusion chromatography was used to evaluate the resultant linear and hyperbranched copolymers’ molecular weight and dispersity, and FT-IR and ¹H-NMR spectroscopy techniques were used to delineate their chemical structure. The structural changes in the obtained self-organized supramolecular structures were thoroughly investigated using aqueous media with varying pH and salinity by dynamic light scattering (DLS), fluorescence spectroscopy (FS), and transmission electron microscopy (TEM). The nanoscale assemblies formed by the amphiphiles indicate significant potential for applications within the field of nanotechnology. Full article

Urban street trees often exhibit distinct architectural characteristics compared to their counterparts in natural forests. Allometric equations for the stem (M_S), branches (M_B), and total dry aboveground biomass of urban trees (M_T) were developed, based on 52 destructively sampled specimens, belonging to 10 different species, growing in the Municipality of Athens, Greece. Linear, log-linear, and nonlinear regression analyses were applied, and fit statistics were used to select the most appropriate model. The results indicated that diameter at breast height (D_1.3) and tree height (H) are needed for accurately predicting M_S, while M_B may be estimated based on D_1.3. To circumvent the caveat of the additivity property for estimating the biomass of different tree component, nonlinear seemingly unrelated regression (NSUR) was implemented. The 95% prediction intervals for M_S, M_B, and M_T efficiently captured the variability of the sampled trees. Finally, the predictions were compared with estimates from i-Tree, the most widely used model suite for urban and rural forestry analysis, and a mean deviation of 134% (ranging from 3% to 520%) was reported. Therefore, in the absence of urban-specific allometries, the obtained empirical models are proposed for estimating biomass in street trees, particularly in cities with Mediterranean-like climatic influences. Full article

Siliceous minerals with the property of resistance to diagenetic alteration precipitate during the migration of hydrothermal fluids through strike-slip faults and the interaction of these fluids with host rocks during fault activity. Based on petrological analyses and U-Pb dating of siliceous minerals, the stages of strike-slip faulting of the ultra-deep-burial Ordovician in the Fuman oilfield were subdivided and their evolutionary process was discussed in combination with seismic interpretation. The results reveal the following: (1) the strike-slip faults contain hydrothermal siliceous minerals, including cryptocrystalline silica, crystalline silica, and radial silica. (2) Based on the twelve U-Pb ages of siliceous minerals (ranging from 458 ± 78 Ma to 174 ± 35 Ma) and five U-Pb ages of calcite, the activity of the strike-slip faults was divided into six stages: the Middle Caledonian, Late Caledonian, Early Hercynian, Middle Hercynian, Late Hercynian, and Yanshanian, corresponding to twelve siliceous U-Pb ages ranging from 458 ± 78 Ma to 174 ± 35 Ma, and five calcitic U-Pb ages. The Late Caledonian and Early Hercynian were the main periods of strike-slip fault activity, while the Late Hercynian period marked the final period of the fault system. (3) Later-stage faults inherited and developed from pre-existing faults. Steep linear strike-slip faults formed during the Middle and Late Caledonian movements. During the Late Hercynian and Yanshanian movements, mid-shallow faults, branch faults, and shallow echelon faults developed on the foundation of these linear faults. The methods and results of this study can guide future hydrocarbon exploration in the Fuman oilfield and can be applied to areas with similar tectonic backgrounds. Full article

Background: Recent studies have demonstrated that branched-chain amino acids are neuroprotective and neurorestorative. Branched-chain amino acid supplements are now being recommended to be taken before contact sports to reduce concussions. While peaks and troughs in branched-chain amino acids have previously been reported in hospital settings, the metabolism of a single recommended dose of over-the-counter branched-chain amino acids has yet to be elucidated. Methods: We analyzed a patented branched-chain amino acid product to assess its metabolism in 10 healthy adults. Results: Over the defined time points, measured levels of branched-chain amino acids remained significantly elevated when compared to the physiological baseline. The elevations in measured plasma levels indicate that a single oral dose is a viable intake option for increasing levels of branched-chain amino acids. Conclusions: This information can be leveraged to better plan branched-chain amino acid-based treatment doses in order to treat pathologies such as brain injury. Full article

Nowadays, object detection algorithms are widely used in various scenarios. However, there are further small object detection requirements in some special scenarios. Due to the problems related to small objects, such as their less available features, unbalanced samples, higher positioning accuracy requirements, and fewer data sets, a small object detection algorithm is more complex than a general object detection algorithm. The detection effect of the model for small objects is not ideal. Therefore, this paper takes YOLOXs as the benchmark network and enhances the feature information on small objects by improving the network’s structure so as to improve the detection effect of the model for small objects. This specific research is presented as follows: Aiming at the problem of a neck network based on an FPN and its variants being prone to information loss in the feature fusion of non-adjacent layers, this paper proposes a feature fusion and distribution module, which replaces the information transmission path, from deep to shallow, in the neck network of YOLOXs. This method first fuses and extracts the feature layers used by the backbone network for prediction to obtain global feature information containing multiple-size objects. Then, the global feature information is distributed to each prediction branch to ensure that the high-level semantic and fine-grained information are more efficiently integrated so as to help the model effectively learn the discriminative information on small objects and classify them correctly. Finally, after testing on the VisDrone2021 dataset, which corresponds to a standard image size of 1080p (1920 × 1080), the resolution of each image is high and the video frame rate contained in the dataset is usually 30 frames/second (fps), with a high resolution in time, it can be used to detect objects of various sizes and for dynamic object detection tasks. And when we integrated the module into a YOLOXs network (named the FE-YOLO network) with the three improvement points of the feature layer, channel number, and maximum pool, the mAP and APs were increased by 1.0% and 0.8%, respectively. Compared with YOLOV5m, YOLOV7-Tiny, FCOS, and other advanced models, it can obtain the best performance. Full article

Hand–object occlusion poses a significant challenge in 3D pose estimation. During hand–object interactions, parts of the hand or object are frequently occluded by the other, making it difficult to extract discriminative features for accurate pose estimation. Traditional methods typically extract features for both the hand and object from a single image using a shared backbone network. However, this approach often results in feature contamination, where hand and object features are mixed, especially in occluded regions. To address these issues, we propose a novel 3D hand–object pose estimation framework that explicitly tackles the problem of occlusion through two key innovations. While existing methods rely on a single backbone for feature extraction, our framework introduces a feature decoupling strategy that shares low-level features (using ResNet-50) to capture interaction contexts, while separating high-level features into two independent branches. This design ensures that hand-specific features and object-specific features are processed separately, reducing feature contamination and improving pose estimation accuracy under occlusion. Recognizing the correlation between the hand’s occluded regions and the object’s geometry, we introduce the Hand–Object Cross-Attention Transformer (HOCAT) module. Unlike traditional attention mechanisms that focus solely on feature correlations, the HOCAT leverages the geometric stability of the object as prior knowledge to guide the reconstruction of occluded hand regions. Specifically, the object features (key/value) provide contextual information to enhance the hand features (query), enabling the model to infer the positions of occluded hand joints based on the object’s known structure. This approach significantly improves the model’s ability to handle complex occlusion scenarios. The experimental results demonstrate that our method achieves significant improvements in hand–object pose estimation tasks on publicly available datasets such as HO3D V2 and Dex-YCB. On the HO3D V2 dataset, the PAMPJPE reaches 9.1 mm, the PAMPVPE is 9.0 mm, and the F-score reaches 95.8%. Full article

The novel plant hormone strigolactones (SL) are involved significantly in plant growth and development. Its key members SMXL6, 7, 8 can modulate SL signal reception and response negatively and can regulate plant branching remarkably. There are relatively scarce studies of cotton SMXL gene family, and this study was carried out to clarify the role of GbSMXL8 in cotton fiber development. Phylogenetic analysis identified 48 cotton SMXL genes, which were divided into SMXL-I (SMXL 1, 2), SMXL-II (SMXL 3) and SMXL-III (SMXL6, 7, 8) groups. The results of the cis-element analysis indicated that the SMXL gene could respond to hormones and the environment to modulate cotton growth process. A candidate gene GbSMXL8 was screened out based on the expression difference in extreme varieties of Gossypium barbadense. Tissue-specific analysis indicated that GbSMXL8 was mainly expressed in roots, 20D, 25D, and 35D and was involved in SL signaling pathways. In vitro ovule culture experiments showed that exogenous SLs (GR24) could promote the fiber elongation of G. barbadense, and GbSMXL8 expression was increased after GR24 treatment, indicating that GbSMXL8 was specifically responsive to GR24 in regulating fiber growth. GbSMXL8 knockout resulted in creased length and number of epidermal hairs and the length of fiber, indicating the interference role of GbSMXL8 gene with the development of cotton fiber. The GbSMXL8 transgenic plant was detected with a higher chlorophyll content and photosynthetic rate than those of the control plant, producing a direct impact on plant growth, yield, and biomass accumulation. GbSMXL8 gene knockout could increase plant height, accelerate growth rate, and lengthen fiber length. Intervening GbSMXL8 may mediate cotton growth, plant type formation and fiber elongation. In conclusion, the present study uncovers the function of GbSMXL8-mediated SL signal in cotton, providing theoretical insight for future breeding of new cotton varieties. Full article