Search

Quantum-inspired neural networks (QNNs) have shown potential in capturing various non-classical phenomena in language understanding, e.g., the emgerent meaning of concept combinations, and represent a leap beyond conventional models in cognitive science. However, there are still two limitations in the existing QNNs: (1) Both storing and invoking the complex-valued embeddings may lead to prohibitively expensive costs in memory consumption and storage space. (2) The use of entangled states can fully capture certain non-classical phenomena, which are described by the tensor product with powerful compression ability. This approach shares many commonalities with the process of word formation from morphemes, but such connection has not been further exploited in the existing work. To mitigate these two limitations, we introduce a Quantum-inspired neural network with Hierarchical Entanglement Embedding (QHEE) based on finer-grained morphemes. Our model leverages the intra-word and inter-word entanglement embeddings to learn a multi-grained semantic representation. The intra-word entanglement embedding is employed to aggregate the constituent morphemes from multiple perspectives, while the inter-word entanglement embedding is utilized to combine different words based on unitary transformation to reveal their non-classical correlations. Both the number of morphemes and the dimensionality of the morpheme embedding vectors are far smaller than the counterparts of words, which would compress embedding parameters efficiently. Experimental results on four benchmark datasets of different downstream tasks show that our model outperforms strong quantum-inspired baselines in terms of effectiveness and compression ability., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

The sympathetic nervous system (SNS) is essential for the body's immediate response to stress, initiating physiological changes that can be measured through sympathetic nerve activity (SNA). While microneurography (MNG) is the gold standard for direct SNA measurement, its invasive nature limits its practical use in clinical settings. This study investigates the use of multi-wavelength photoplethysmography (PPG) as a non-invasive alternative for SNA measurement. Key features are extracted from the pulsatile components of red and green PPG signals to train a linear regression machine learning (ML) model to predict R-wave-triggered spike count (SPR), a biomarker derived from MNG. The study correlates PPG-derived features with ground truth SPR to develop a predictive model capable of detecting SNA during induced physical stress (isometric handgrip and cold pressor) and cognitive stress (mental arithmetic and Stroop test). Unlike previous research that relies on subjective stress indicators, our work utilizes MNG-derived SPR as an objective ground truth for validation. Our findings demonstrate strong agreement between PPG-predicted SPR values and those obtained via MNG, with red PPG showing a higher correlation. The green wavelength PPG exhibits greater sensitivity in detecting stress-induced SNA, particularly during stress onset, where it outperforms the MNG method in capturing immediate responses to stressors such as mental arithmetic and the cold pressor task. To the best of our knowledge, this is the first study to directly compare PPG-derived SNA estimates with MNG, offering a promising pathway for developing wearable, non-invasive tools for continuous stress monitoring and sympathetic arousal detection., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024. Published by Elsevier Ltd.)

Objective: We investigated the characteristics of hip, knee, and ankle joint reaction forces (JRFs) in stroke patients with spastic hemiplegia during sit-to-stand (Si-St) and stand-to-sit (St-Si) movements and explored the relationship between JRFs and joint moments., Methods: Thirteen stroke patients with spastic hemiplegia and thirteen age-matched healthy subjects were recruited in this study. Three-dimensional motion capture system and force plates were employed to collect kinematic data and ground reaction forces during Si-St and St-Si tasks. Musculoskeletal modeling was used to calculate JRFs and joint moments of the hip, knee and ankle joints during these movements., Results: Those stroke patients exhibited higher JRFs (ranging from 13.49 N/kg to 43.18 N/kg) compared to the healthy group (ranging from 15.81 N/kg to 23.00 N/kg) in hip and knee joints. Furthermore, hip and ankle JRFs exhibited positive correlations with their joint moment in stroke patients during Si-St and St-Si movements., Conclusion: Early rehabilitation strategies should focus on correcting weight distribution and enhancing symmetry in joint loading. Interventions focused on reducing joint moment, particularly for hip and ankle joints, may prove effective in mitigating lower extremity JRFs in stroke patients with spastic hemiplegia during Si-St and St-Si movements., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2025 Elsevier Ltd. All rights reserved.)

Squamous cell carcinoma is a prevalent cancer type that affects various organs in the human body. Manual analysis for detecting squamous cell carcinoma in histopathological images is time-consuming and may be subjective. Squamous cell carcinoma diagnosis is typically based on the differences in the architectural arrangement of squamous epithelial layers and the presence of keratinization. However, the existing literature has predominantly concentrated on identifying cellular irregularities with high magnification images and considering specific organs of squamous cell carcinoma origin. In contrast, relatively little attention has been given to recognizing structural abnormalities observable at low magnification images. In this paper, we consider squamous cell carcinoma histopathological images across different organs of origin captured at low magnification and these images are gathered from various centers to develop a robust model. We propose a novel deep neural network model (MultiSCCHisto-Net) that can detect squamous cell carcinoma of any organ irrespective of organ of origin. In addition, deep neural networks used for histopathological image analysis typically have many parameters, making them computationally expensive. To address this research gap, we incorporate knowledge distillation, which compresses knowledge from a complex teacher model (MultiSCCHisto-Net) into a smaller student model (MultiSCCHisto-Net-KD) while preserving performance and enhancing the generalization of the student model by learning from the teacher's intermediate layer features. Moreover, an explainable deep learning technique called gradient-weighted class activation mapping is incorporated to highlight the image areas that help to classify the sample into particular classes. This explainability significantly enhances our confidence in the proposed model outcomes. We evaluate the model's robustness using private multi-centric and publicly available datasets. Our results show that accuracy rates of 97% and 93% are achieved on private and public datasets, respectively, surpassing the performance of state-of-the-art models., Competing Interests: Declaration of competing interest The authors have no conflicts of interest to declare. All co-authors have seen and agree with the contents of the manuscript and there is no financial interest to report. We certify that the submission is original work and is not under review at any other publication., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)

A rational evaluation of the rural credit environment's current state and identification of its critical issues are crucial for enhancing the construction of rural social credit systems. Drawing on the "Outline for the Construction of the Social Credit System (2014-2020)" and related literature, this paper clarifies the concepts and measurement principles of the rural credit environment. This study innovatively constructs an evaluation framework for the rural credit environment and conducts quantitative measurements and statistical analyses using a combined weighting method to delineate the environment's current state. Using Banqiao Town as a case study, this paper employs the cloud model to evaluate the rural credit environment's quality levels and to identify key factors influencing its quality effectively. From a macro perspective, statistical measurements and index analyses establish evaluation standards and quantify the state of the rural credit environment. In contrast, from a micro perspective, case analysis focuses on evaluating the quality levels and identifying crucial issues in specific locales. Integrating macro and micro perspectives offers a novel approach to evaluating the rural credit environment, offering theoretical approaches and practical strategies for promoting a favorable credit environment and advancing the construction of rural social credit systems in China., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper, (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Pollen development and germination are critical for successful generation of offspring in plants, yet they are highly susceptible to heat stress (HS). However, the molecular mechanism underlying this process has not been fully elucidated. In this study, we highlight the essential roles of two mRNA capping enzymes, named Arabidopsis mRNA capping phosphatase (ARCP) 1 and 2, in regulating male and female gamete development. The transmission efficiencies of gametes carrying arcp1 arcp2 from arcp1 +/- arcp2 -/- and arcp1 -/- arcp2 +/- mutants are 30% and zero, respectively. These mutants exhibited a significant increase in misshaped pollen, with germination rates approximately half of those in wild type. ARCP1/2 exhibit RNA triphosphatase and RNA guanylyltransferase activities, which are required for proper pollen development. Through RNA-seq analysis, genes involved in pollen development/germination and HS response were identified as downregulated genes in pollen from arcp1 +/- arcp2 -/- mutant. Furthermore, ARCP2 protein is degraded under HS condition, and inducing the expression of ARCP2 can increase the pollen germination rate under elevated temperature. We propose that HS triggers the degradation of mRNA capping enzymes, which in turn disrupts the transcriptome that required for pollen development and pollen germination and ultimately leads to male sterility., (© 2024 John Wiley & Sons Ltd.)

Identification of potential drug-target interactions (DTIs) is a crucial step in drug discovery and repurposing. Although deep learning effectively deciphers DTIs, most deep learning-based methods represent drug features from only a single perspective. Moreover, the fusion method of drug and protein features needs further refinement. To address the above two problems, in this study, we develop a novel end-to-end framework named DO-GMA for potential DTI identification by incorporating D epthwise O verparameterized convolutional neural network and the G ated M ultihead A ttention mechanism with shared-learned queries and bilinear model concatenation. DO-GMA first designs a depthwise overparameterized convolutional neural network to learn drug representations from their SMILES strings and protein representations from their amino acid sequences. Next, it extracts drug representations from their 2D molecular graphs through a graph convolutional network. Subsequently, it fuses drug and protein features by combining the gated attention mechanism and the multihead attention mechanism with shared-learned queries and bilinear model concatenation. Finally, it takes the fused drug-target features as inputs and builds a multilayer perceptron to classify unlabeled drug-target pairs (DTPs). DO-GMA was benchmarked against six newest DTI prediction methods (CPI-GNN, BACPI, CPGL, DrugBAN, BINDTI, and FOTF-CPI) under four different experimental settings on four DTI data sets (i.e., DrugBank, BioSNAP, C.elegans, and BindingDB). The results show that DO-GMA significantly outperformed the above six methods based on AUC, AUPR, accuracy, F1-score, and MCC. An ablation study, robust statistical analysis, sensitivity analysis of parameters, visualization of the fused features, computational cost analysis, and case analysis further validated the powerful DTI identification performance of DO-GMA. In addition, DO-GMA predicted that two drug-protein pairs (i.e., DB00568 and P06276, and DB09118 and Q9UQD0) could be interacting. DO-GMA is freely available at https://github.com/plhhnu/DO-GMA.

How to raise donations effectively, especially in the E-era, has puzzled fundraisers and scientists across various disciplines. Our research focuses on donation-based crowdfunding projects and investigates how the emotional valence expressed verbally (in textual descriptions) and visually (in facial images) in project descriptions affects project performance. Study 1 uses field data (N = 3817), grabs project information and descriptions from a top donation-based crowdfunding platform, computes visual and verbal emotional valence using a deep-learning-based affective computing method and analyses how multimodal emotional valence influences donation outcomes. Study 2 conducts experiments in GPT-4 (Study 2a, N = 400) and humans (Study 2b, N = 240), manipulates the project's visual and verbal emotional valence through AI-generated stimuli and then assesses donation decisions (both GPT-4 and humans) and corresponding state empathy (humans). The results indicate a multimodal positivity superiority effect: both visual and verbal emotional valence promote initial whether-to-donate decisions, whereas only verbal emotional valence further promotes the how-much-to-donate decisions. Notably, such multimodal emotional effects can be explained through different mediating paths of empathic concern and empathic hopefulness. The current study theoretically facilitates our understanding of the emotional motivations underlying human prosociality and provides insights into crafting impactful advertisements for online donations., (© 2025 The British Psychological Society.)

Diabetes mellitus (DM) represents a major global health challenge, affecting a diverse range of demographic populations across all age groups. It has particular implications for women during pregnancy and the postpartum period. The contemporary prevalence of sedentary lifestyle patterns and suboptimal dietary practices has substantially contributed to the escalating incidence of this metabolic disorder. The timely identification of diabetes mellitus (DM) in the female population is crucial for preventing related complications and facilitating the implementation of effective therapeutic interventions. However, conventional predictive models frequently demonstrate limited external validity when applied across heterogeneous datasets, potentially compromising clinical utility. This study proposes a robust machine learning (ML) framework for diabetes prediction across diverse populations using two distinct datasets: the PIMA and BD datasets. The framework employs intra-dataset, inter-dataset, and partial fusion dataset validation techniques to comprehensively assess the generalizability and performance of various models. In intra-dataset validation, the Extreme Gradient Boosting (XGBoost) model achieved the highest accuracy on the PIMA dataset with 79%. In contrast, the Random Forest (RF) and Gradient Boosting (GB) models demonstrated accuracy close to 99% on the BD dataset. For inter-dataset validation, where models were trained on one dataset and tested on the other, the ensemble model outperformed others with 88% accuracy when trained on PIMA and tested on BD. However, model performance declined when trained on BD and tested on PIMA (74%), reflecting the challenges of inter-dataset generalization ability. Finally, during partial fusion data validation, the deep learning (DL) model achieved 74% accuracy when trained on the BD dataset augmented with 30% of the PIMA dataset. This accuracy increased to 98% when training on the PIMA dataset combined with 30% of the BD data. These findings emphasize the importance of dataset diversity and the partial fusion dataset that can significantly enhance the model's robustness and generalizability. This framework offers valuable insights into the complexities of diabetes prediction across heterogeneous environments., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2025 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by social communication and interaction problems. The prevalence of ASD is increasing globally, with a higher ratio of males to females. Gastrointestinal symptoms are common in individuals with ASD, and gut microbiota has been implicated in the disorder's development. This study aimed to investigate the gut microbiota alteration in Thai individuals with ASD compared to healthy controls using 16S rRNA gene sequencing. The influence of gender and age on gut microbiota composition and function was also examined. A total of 65 ASD individuals and 30 neurotypical (NT) individuals were included in the analysis. The results revealed notable differences in gut microbiota composition between the ASD and NT groups, with variations observed in microbial richness and the presence of enriched microbial taxa. These differences were influenced by both gender and age. Fusobacteriota, Fusobacteriaceae, and Fusobacterium were found to be enriched in individuals with ASD. Furthermore, the study identified gender-related taxa, such as Bacteroides plebeius, enriched in ASD females. Age-related taxa, including Veillonella, known to be associated with poor oral hygiene, were also observed in ASD children. The analysis of differentially abundant pathways highlighted the enrichment of various metabolic pathways in individuals with ASD, including those related to endocrine-disrupting chemicals. These findings underscore the importance of considering gender and age when studying gut microbiota in ASD. They provide valuable insights into the potential role of gut microbiota dysbiosis in ASD pathogenesis and highlight the influence of environmental factors., Competing Interests: Declarations. Competing interests: The authors declare no competing interests., (© 2025. The Author(s).)

Background: Short-form video applications have spread rapidly and gained popularity among adolescents for their interactivity, sociality, and personalization. However, excessive and uncontrolled use has also caused some adolescents to develop short-form video addiction (SFVA). Although parental phubbing has been shown to be a risk factor for SFVA in adolescents, its intermediate mechanisms, particularly the emotional mediation mechanisms, remain unclear. Therefore, this study aims to explore the relationships among parental phubbing, SFVA, symptoms of depression and anxiety, and neuroticism in adolescents., Methods: Adolescents aged 13-19 (N = 5785) were recruited from Shandong Province, China, as participants in 2023. Participants were asked to complete five self-report questionnaires, including the Parental Phubbing Behavior Questionnaire (PPBQ), Short-Form Video Addiction (SFVA) Scale, The Center for Epidemiological Studies Depression Scale (CESD20), Generalized Anxiety Disorder 7-item (GAD-7) Scale, and Chinese Big Five Personality Inventory (Brief Version) (CBF-PI-B)., Results: The results indicate that the influence of parental phubbing on short-form video addiction among adolescents is mediated by symptoms of depression and anxiety. Additionally, neuroticism moderates the association between parental phubbing and symptoms of depression and anxiety., Limitations: Cross-sectional design limits causal inferences. Reliance on self-report scales., Conclusions: These findings contribute to a better understanding of the systemic impact mechanisms of negative family interactions on SFVA. Moreover, prevention and intervention strategies targeting high neuroticism and symptoms of depression and anxiety may help prevent the developmental pathway from parental phubbing to SFVA., Competing Interests: Declaration of competing interest None., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Introduction: Weeds are a major factor affecting crop yield and quality. Accurate identification and localization of crops and weeds are essential for achieving automated weed management in precision agriculture, especially given the challenges in recognition accuracy and real-time processing in complex field environments. To address this issue, this paper proposes an efficient crop-weed segmentation model based on an improved UNet architecture and attention mechanisms to enhance both recognition accuracy and processing speed., Methods: The model adopts the encoder-decoder structure of UNet, utilizing MaxViT (Multi-Axis Vision Transformer) as the encoder to capture both global and local features within images. Additionally, CBAM (Convolutional Block Attention Module) is incorporated into the decoder as a multi-scale feature fusion module, adaptively adjusting feature map weights to enable the model to focus more accurately on the edges and textures of crops and weeds., Results and Discussion: Experimental results show that the proposed model achieved 84.28% mIoU and 88.59% mPA on the sugar beet dataset, representing improvements of 3.08% and 3.15% over the baseline UNet model, respectively, and outperforming mainstream models such as FCN, PSPNet, SegFormer, DeepLabv3+, and HRNet. Moreover, the model's inference time is only 0.0559 seconds, reducing computational overhead while maintaining high accuracy. Its performance on a sunflower dataset further verifies the model's generalizability and robustness. This study, therefore, provides an efficient and accurate solution for crop-weed segmentation, laying a foundation for future research on automated crop and weed identification., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2025 Li, Guo, Li, Ji, Wu, Chen, Han, Han, Liu, Ruan and Yang.)

Long COVID and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) patients share similar symptoms including post-exertional malaise, neurocognitive impairment, and memory loss. The neurocognitive impairment in both conditions might be linked to alterations in the hippocampal subfields. Therefore, this study compared alterations in hippocampal subfields of 17 long COVID, 29 ME/CFS patients, and 15 healthy controls (HC). Structural MRI data was acquired with sub-millimeter isotropic resolution on a 7 Telsa MRI scanner and hippocampal subfield volumes were then estimated for each participant using FreeSurfer software. Our study found significantly larger volumes in the left hippocampal subfields of both long COVID and ME/CFS patients compared to HC. These included the left subiculum head (long COVID; p = 0.01, ME/CFS; p = 0.002,), presubiculum head (long COVID; p = 0.004, ME/CFS; p = 0.005), molecular layer hippocampus head (long COVID; p = 0.014, ME/CFS; p = 0.011), and whole hippocampal head (long COVID; p = 0.01, ME/CFS; p = 0.01). Notably, hippocampal subfield volumes were similar between long COVID and ME/CFS patients. Additionally, we found significant associations between hippocampal subfield volumes and severity measures of 'Pain', 'Duration of illness', 'Severity of fatigue', 'Impaired concentration', 'Unrefreshing sleep', and 'Physical function' in both conditions. These findings suggest that hippocampal alterations may contribute to the neurocognitive impairment experienced by long COVID and ME/CFS patients. Furthermore, our study highlights similarities between these two conditions., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2025 Thapaliya et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Introduction: Ultra-high-field magnetic resonance (MR) systems (7 T and 9.4 T) offer the ability to probe human brain metabolism with enhanced precision. Here, we present the preliminary findings from 3D MR spectroscopic imaging (MRSI) of the human brain conducted with the world's first 10.5 T whole-body MR system., Methods: Employing a custom-built 16-channel transmit and 80-channel receive MR coil at 10.5 T, we conducted MRSI acquisitions in six healthy volunteers to map metabolic compounds in the human cerebrum in vivo. Three MRSI protocols with different matrix sizes and scan times (4.4 × 4.4 × 4.4 mm³: 10 min, 3.4 × 3.4 × 3.4 mm³: 15 min, and 2.75×2.75×2.75 mm³: 25 min) were tested. Concentric ring trajectories were utilized for time-efficient encoding of a spherical 3D k-space with ∼4 kHz spectral bandwidth. B 0 /B 1 shimming was performed based on respective field mapping sequences and anatomical T 1 -weighted MRI were obtained., Results: By combining the benefits of an ultra-high-field system with the advantages of free-induction-decay (FID-)MRSI, we present the first metabolic maps acquired at 10.5 T in the healthy human brain at both high (voxel size of 4.4³ mm³) and ultra-high (voxel size of 2.75³ mm³) isotropic spatial resolutions. Maps of 13 metabolic compounds (aspartate, choline compounds and creatine + phosphocreatine, γ-aminobutyric acid (GABA), glucose, glutamine, glutamate, glutathione, myo-inositol, scyllo-inositol, N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), taurine) and macromolecules were obtained individually. The spectral quality was outstanding in the parietal and occipital lobes, but lower in other brain regions such as the temporal and frontal lobes. The average total NAA (tNAA = NAA + NAAG) signal-to-noise ratio over the whole volume of interest was 12.1± 8.9 and the full width at half maximum of tNAA was 24.7± 9.6 Hz for the 2.75 × 2.75 × 2.75 mm³ resolution. The need for an increased spectral bandwidth in combination with spatio-spectral encoding imposed significant challenges on the gradient system, but the FID approach proved very robust to field inhomogeneities of ∆B 0 = 45 ± 38 Hz (frequency offset ± spatial STD) and B 1 = 65 ± 11° within the MRSI volume of interest.+ = 65 ± 11° within the MRSI volume of interest., Discussion: These preliminary findings highlight the potential of 10.5 T MRSI as a powerful imaging tool for probing cerebral metabolism. By providing unprecedented spatial and spectral resolution, this technology could offer a unique view into the metabolic intricacies of the human brain, but further technical developments will be necessary to optimize data quality and fully leverage the capabilities of 10.5 T MRSI., Competing Interests: Declaration of competing interest No author has any financial interest to declare., (Copyright © 2025 The Author(s). Published by Elsevier Inc. All rights reserved.)

Dexter energy transfer (DET) of triplet electronic states is used to direct energy in photovoltaics, quench reactive singlet oxygen species in biological systems, and generate them in photodynamic therapy. However, the extent to which repeated DET between aromatic residues can lead to triplet energy migration in proteins has not been investigated. Here, we computationally describe DET rates in microtubules, actin filaments and the intermediate filament, vimentin. We discover instances where interaromatic residue Dexter couplings within individual protein subunits of these polymers are similar those of small molecules used for organic electronics. However, interaromatic residue coupling is mostly weak (<10 -3 eV), limiting triplet energy diffusion lengths to 6.1, 0.5 and 1.0 Å in microtubules, actin filaments and vimentin, respectively. On the other hand, repeated förster resonance energy transfer (FRET) between aromatic residues leads to singlet energy diffusion lengths of 12.4 Å for actin filaments and about 8.6 Å for both microtubules and vimentin filaments. Our work shows that singlet energy migration dominates over triplet energy migration in cytoskeletal polymers.

Carbon emissions have increasingly been the focus of all governments as countries throughout the world choose carbon neutrality as a national development strategy. The analysis of the spatiotemporal dynamics of CO 2 emission has emerged as a significant research topic considering the dual-carbon goal. In this research, we explore the spatiotemporal changes of CO 2 emission at different scales based on nighttime light data. The Chinese Academy of Science's Earth Luminous Dataset, CO 2 emission data from Carbon Emission Accounts and Datasets, and basic national geographical data are used for analysis. A linear regression model between nighttime light data and CO 2 emission is constructed. Thereafter, the global Moran's I index of exploratory spatial data analysis is used to verify the spatial parameters of all provinces. The trend value method is utilized to analyze the changing trend of CO 2 emission at multiscale levels, covering the Chinese mainland, three major economic regions, and six largest agglomerations from 2012 to 2019. Experimental results show a significant positive correlation between the CO 2 emission and nighttime light data from 2012 to 2019. The nighttime light data could be used to effectively estimate the total CO 2 emission at the provincial and municipal levels in China. The growth rate of CO 2 emissions in China is stable and has decreased in 2015. Furthermore, the spatiotemporal dynamics of CO 2 emission in different agglomerations vary. Our work provides a scientific basis for the different provinces and cities to develop feasible emission reduction policies., Competing Interests: The authors declare that there is no conflict of interests regarding the publication of this article., (© 2025 The Authors.)

With the development of electronic technology, triboelectric-based sensors have been widely researched in fields such as healthcare, rehabilitation training, and sports assistance due to their manufacturing convenience and self-powering advantages. Among them, 3D fabric-based triboelectric sensors not only possess advantages such as easy mechanized production, good breathability, and ease of wearing but also their unique 3D structure enhances the specific surface area, thereby amplifying the sensitivity. This study proposes a 3D bristle-structured fabric made by a digital knitting technology that has not been studied widely for triboelectric devices. By applying the 3D bristle structure with a large specific surface area to the single jersey fabric, the effective contact area during friction can be increased, resulting in a higher surface charge density. Additionally, the microcapacitor-like effect provided by the numerous microstructures allows the device to store more surface charge, further improving the output performance. The study systematically investigates the output performance of four different structures assembled by single jersey and 3D bristle-structured fabrics. The optimal sample exhibits a 57% higher output voltage than that of the reference 2D fabric sample. The 3D bristle-structured fabric demonstrates linear high sensitivity and distinct output performance when used as a sensor. Finally, a machine learning integration is applied to judge motion to assist a baseball pitcher in a self-training system.

Mycelium-based composites hold significant potential as sustainable alternatives to traditional materials, offering innovative solutions to the escalating challenges of global warming and climate change. This review examines their production techniques, advantages, and limitations, emphasizing their role in addressing pressing environmental and economic concerns. Current applications span various industries, including manufacturing and biomedical fields, where mycelium-based composites demonstrate the capacity to mitigate environmental impact and enhance economic sustainability. Key findings highlight their environmental benefits, economic viability, and versatile applications, showcasing their potential to revolutionize multiple sectors. However, challenges such as consumer acceptance, intrinsic variability, and the need for standardized guidelines persist, underscoring the importance of further research and innovation. By optimizing material properties and refining production processes, mycelium-based composites could pave the way for widespread adoption as sustainable materials, contributing to a greener and more environmentally conscious future., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2025 The Authors. Published by Elsevier Inc. All rights reserved.)

Despite the expanding therapeutic options available to cancer patients, therapeutic resistance, disease recurrence, and metastasis persist as hallmark challenges in the treatment of cancer. The rise to prominence of generative artificial intelligence (GenAI) in many realms of human activities is compelling the consideration of its capabilities as a potential lever to advance the development of effective cancer treatments. This article presents a hypothetical case study on the application of generative pre-trained transformers (GPTs) to the treatment of metastatic prostate cancer (mPC). The case explores the design of GPT-supported adaptive intermittent therapy for mPC. Testosterone and prostate-specific antigen (PSA) are assumed to be repeatedly monitored while treatment may involve a combination of androgen deprivation therapy (ADT), androgen receptor-signalling inhibitors (ARSI), chemotherapy, and radiotherapy. The analysis covers various questions relevant to the configuration, training, and inferencing of GPTs for the case of mPC treatment with a particular attention to risk mitigation regarding the hallucination problem and its implications to clinical integration of GenAI technologies. The case study provides elements of an actionable pathway to the realization of GenAI-assisted adaptive treatment of metastatic prostate cancer. As such, the study is expected to help facilitate the design of clinical trials of GenAI-supported cancer treatments., Competing Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article., (© The Author(s) 2025.)

This study introduces a novel AI-driven approach to support elderly patients in Thailand with medication management, focusing on accurate drug label interpretation. Two model architectures were explored: a Two-Stage Optical Character Recognition (OCR) and Large Language Model (LLM) pipeline combining EasyOCR with Qwen2-72b-instruct and a Uni-Stage Visual Question Answering (VQA) model using Qwen2-72b-VL. Both models operated in a zero-shot capacity, utilizing Retrieval-Augmented Generation (RAG) with DrugBank references to ensure contextual relevance and accuracy. Performance was evaluated on a dataset of 100 diverse prescription labels from Thai healthcare facilities, using RAG Assessment (RAGAs) metrics to assess Context Recall, Factual Correctness, Faithfulness, and Semantic Similarity. The Two-Stage model achieved high accuracy (94%) and strong RAGAs scores, particularly in Context Recall (0.88) and Semantic Similarity (0.91), making it well-suited for complex medication instructions. In contrast, the Uni-Stage model delivered faster response times, making it practical for high-volume environments such as pharmacies. This study demonstrates the potential of zero-shot AI models in addressing medication management challenges for the elderly by providing clear, accurate, and contextually relevant label interpretations. The findings underscore the adaptability of AI in healthcare, balancing accuracy and efficiency to meet various real-world needs.

Because of its dimensional characteristics, two-dimensional (2D) materials exhibit many special properties. The key to researching their features is to prepare high-quality larger-area monolayer 2D materials. Metal-assisted mechanical exfoliation method offers the possibility. However, there is still an unclear mechanism. We take transition metal dichalcogenides (TMD) as examples to perform serials of first-principles calculations to study their interlayer interaction. The calculated results demonstrate that the main factors affecting the interlayer interaction of 2D TMD materials include their structural features and the elemental features. The smaller interlayer distance between the topmost two layers than that of bulk leads to the stronger interlayer bonding energy, which results in the difficulty to exfoliate the large-area monolayer 2D crystals successfully. Further, we selected Ti, Cr and Au as the assisted metal due to their enhanced density of states near the Fermi level to calculate the adhesion energies of TMD on their surfaces. We found that there exists strong adhesion energy between 2D crystal and metal, which stretches the spacing of the topmost two layers and leads to the electron density rearrangement. Therefore, metal assisted mechanical exfoliation facilitates the successful preparation of monolayer 2D crystal., Competing Interests: Declarations. Competing interests: The authors declare no competing interests., (© 2025. The Author(s).)

Introduction: Stroke is a major cause of morbidity and mortality worldwide. While electroencephalography (EEG) offers valuable data on post-stroke brain activity, qualitative EEG assessments may be misinterpreted. Therefore, we examined the potential of quantitative EEG (qEEG) to identify key band frequencies that could serve as potential electrophysiological biomarkers in stroke patients., Materials and Methods: A single-centre case-control study was conducted in which patients admitted with stroke and healthy controls were recruited with consent. EEG was performed within 48 hours of admission for stroke patients and during outpatient assessments for controls. The EEG signals were pre-processed, analysed for spectral power using MATLAB, and plotted as topoplots., Results: A total of 194 participants were included and equally divided into patients with ischemic stroke and controls. The mean age of our study cohort was 55.11 years (SD±13.12), with a median National Institute of Health Stroke Scale (NIHSS) score of 6 (IQR 4-6) and lacunar stroke was the most common subtype (49.5%). Spectral analysis, with subsequent topographic brain mapping, highlighted clustering of important channels within the beta, alpha, and gamma bands., Conclusion: qEEG analysis identified significant band frequencies of interest in post-stroke patients, suggesting a role as a diagnostic and prognostic tool. Topographic brain mapping provides a precise representation that can guide interventions and rehabilitation strategies. Future research should explore the use of machine learning for stroke detection and provide individualized treatment.

Flavonoids are plant-derived compounds that have several health benefits, including antioxidative, anti-inflammatory, anti-mutagenic, and anti-carcinogenic effects. Quercetin is a flavonoid that is widely present in various fruits, vegetables, and drinks. Accurate determination of quercetin in different samples is of great importance for its potential health benefits. This review, is an overview of sample preparation and determination methods for quercetin in diverse matrices. Previous research on sample preparation and determination methods for quercetin are summarized, highlighting the advantages and disadvantages of each method and providing insights into recent developments in quercetin sample treatment. Various analytical techniques are discussed including spectroscopic, chromatographic, electrophoretic, and electrochemical methods for the determination of quercetin and its derivatives in different samples. UV-Vis (Ultraviolet-visible) spectrophotometry is simple and inexpensive but lacks selectivity. Chromatographic techniques (HPLC, GC) offer selectivity and sensitivity, while electrophoretic and electrochemical methods provide high resolution and low detection limits, respectively. The aim of this review is to comprehensively explore the determination methods for quercetin and quercetin glycosides in diverse matrices, with emphasis on pharmaceutical and biological samples. The review also provides a theoretical basis for method development and application for the analysis of quercetin and quercetin glycosides in real samples.

Importance: Resilience has been reported as a vital element against the high burden on caregivers of children with autism. However, the intricate mechanisms underlying the concurrent interplay of positive factors with resilience within this demographic group remain less understood., Objective: To construct a model to verify the sequential mediating influence of hope and gratitude in the relationship between perceived social support and resilience among caregivers of children with autism., Design: Cross-sectional study., Participants: A total of 193 caregivers of children with autism in mainland China., Outcomes and Measures: Data were collected using the Herth Hope Index, C-GQ-6 Gratitude Questionnaire, Perceived Social Support Scale, and Resilience Questionnaire for Parents of Children With Special Needs. The hypothesized mediating model was tested with Process 4.0 (Model 6) for SPSS (Version 26.0)., Results: The relationship of perceived social support and resilience was mediated by hope, gratitude, and the hope-gratitude sequence. Hope exhibited a stronger mediation effect, accounting for 26.23% of the total effect., Conclusions and Relevance: The results indicated significant positive correlations between perceived social support, hope, gratitude, and resilience, with perceived social support strongly linked to the other three. This study has important implications for occupational therapy, suggesting strategies for resilience-focused interventions tailored to caregivers of children with autism., Plain-Language Summary: Resilience is essential for caregivers of children with autism. Studies show that parents of children with autism report having fewer social supports, reduced quality of life, and heightened mental health challenges, including anxiety, depression, and hopelessness. This research highlights the connections between caregivers' perceived social support, hope, gratitude, and resilience. Addressing caregiver social supports and mental health can enhance caregiver well-being as well as children's participation in meaningful activities. The findings suggest practical implications and insights for occupational therapy practitioners to strengthen and foster caregiver resilience and well-being in the autism community., (Copyright © 2025 by the American Occupational Therapy Association, Inc.)

Fear overgeneralization is widely considered to be a core feature of pathological anxiety, and the excessive spread of fear can be a great burden on patients. Although perceptual processing is a basic process of human cognition, the impact of visual perception recognition on fear generalization has not yet been fully explored. In this study, 69 subjects were recruited to explore the effects of the hierarchical features of stimuli on fear generalization and examine the roles of trait anxiety (TA) and intolerance of uncertainty (IU) in this process. The results reveal that compared with the local features, the subjects showed stronger fear responses to global features of the stimuli, and the weakest fear response was to the reversed features. Furthermore, the results preliminarily suggest that in generalization dominated by global features, TA intensifies the fear response; while in generalization dominated by local features, along with TA, IU makes a unique contribution to increasing the magnitude of fear generalization. These results confirm new potential pathways for fear generalization, specifically the configurational hierarchical features of stimuli. This is explained through object recognition and cognitive bias, implying that cognitive processes and personality traits have interacting effects on fear generalization., Competing Interests: Declaration of competing interest None., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

With the increasing focus on triboelectric-based sensors, research on synthesizing dielectric layers from specific substances is gradually emerging. Despite numerous negatively-charged triboelectric materials, there is a scarcity of synthesizable positively-charged materials, creating a research gap. This study demonstrates the molecular design of a conjugated, mesoporous, self-assembled sheet via bottom-up synthesis. The synthesized sheet is functionalized to create a triboelectric nanogenerator. Its large specific surface area, softness, and internal space increase the actual contact area and provide adsorption sites for polypyrrole nanoparticles. The incorporation of -COO functional group enhances positive triboelectric performance, forming a dielectric layer with charge-trapping capabilities. When contact with polytetrafluoroethylene (PTFE), this structure boosts the output voltage, showing significant amplification after charge injection with minimal decay. As a demonstration, the bilayer structure is applied as a touchpad on the experimenter's arm to write symbols. The signals are input into an innovative machine-learning model to interpret the writer's intent. Additionally, the device connects to a terminal for real-time medical services, suggesting practical applications for wearable triboelectric sensors with artificial intelligence., (© 2024 Wiley‐VCH GmbH.)

In Magnetic Resonance Imaging (MRI), the sequential acquisition of raw complex-valued image data in Fourier space, also known as k-space, results in extended examination times. To speed up the MRI scans, k-space data are usually undersampled and processed using numerical techniques such as compressed sensing (CS). While the majority of CS-MRI algorithms primarily focus on magnitude images due to their significant diagnostic value, the phase components of complex-valued MRI images also hold substantial importance for clinical diagnosis, including neurodegenerative diseases. In this work, complex-valued MRI reconstruction is studied with a focus on the simultaneous reconstruction of both magnitude and phase images. The proposed algorithm is based on the nonsubsampled contourlet transform (NSCT) technique, which offers shift invariance in images. Instead of directly transforming the complex-valued image into the NSCT domain, we introduce a wavelet transform within the NSCT domain, reducing the size of the sparsity of coefficients. This two-level hierarchical constraint (HC) enforces sparse representation of complex-valued images for CS-MRI implementation. The proposed HC is seamlessly integrated into a proximal algorithm simultaneously. Additionally, to effectively minimize the artifacts caused by sub-sampling, thresholds related to different sub-bands in the HC are applied through an alternating optimization process. Experimental results show that the novel method outperforms existing CS-MRI techniques in phase-regularized complex-valued image reconstructions., Competing Interests: Declaration of competing Interest The authors declare no conflict of interest., (Copyright © 2024. Published by Elsevier Inc.)

The majority of patients hospitalised for trauma survive their injuries, with the quality of the survival potentially influenced by early acute hospital rehabilitation. The aim of this study was to review the outcomes of patients managed under an intensive Allied Health Model of Care (AHMOC) compared to a baseline cohort., Methods: The AHMOC was commenced in February 2020 on the Trauma ward at Alfred Health for 12 months. The baseline group included patients admitted to the trauma ward in 2019. All patients who were registered by either the Victorian Orthopaedic Trauma Outcomes Registry (VOTOR) or the Victorian State Trauma Registry (VSTR). The association between the groups (AHMOC and baseline) and outcomes were assessed using logistic (discharge destination, 12-month return to work) and linear (length of stay (LOS)) regression., Results: There were 1644 in the baseline group and 1732 in the AHMOC group, of which 70 % and 76 % respectively were discharged directly home from the acute hospital (p value < 0.001). The trauma ward LOS was 4.0 days for each group (p value 0.77). After accounting for confounders, the adjusted odds of discharge home and RTW at 12 months were 53 % (AOR 1.53 95 % CI 1.29, 1.82) and 65 % (AOR 1.65 95% CI 1.24, 2.21) higher for the AHMOC group compared to baseline, respectively. There was also a 6 % reduction in the LOS in the AHMOC group compared to baseline (Adjusted mean difference 6 %; 95 % CI (0.881, 0.999) p value = 0.050)., Conclusion: This study shows an association between an AHMOC for trauma patients with a focus on early and intensive therapy and improved odds of discharge directly home, as well as improved 12 month return to work outcomes., Competing Interests: Declaration of competing interest None., (Copyright © 2024 The Author(s). Published by Elsevier Ltd.. All rights reserved.)

Predicting drug-drug interactions (DDIs) is crucial for understanding and preventing adverse drug reactions (ADRs). However, most existing methods inadequately explore the interactive information between drugs in a self-supervised manner, limiting our comprehension of drug-drug associations. This paper introduces EDDINet: Enhancing Drug-Drug Interaction Prediction via Information Flow and Consensus-Constrained Multi-Graph Contrastive Learning for precise DDI prediction. We first present a cross-modal information-flow mechanism to integrate diverse drug features, enriching the structural insights conveyed by the drug feature vector. Next, we employ contrastive learning to filter various biological networks, enhancing the model's robustness. Additionally, we propose a consensus regularization framework that collaboratively trains multi-view models, producing high-quality drug representations. To unify drug representations derived from different biological information, we utilize an attention mechanism for DDI prediction. Extensive experiments demonstrate that EDDINet surpasses state-of-the-art unsupervised models and outperforms some supervised baseline models in DDI prediction tasks. Our approach shows significant advantages and holds promising potential for advancing DDI research and improving drug safety assessments. Our codes are available at: https://github.com/95LY/EDDINet&#95;code., Competing Interests: Declaration of competing interest The authors declare no conflict of interests., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)