Search Results (10,151)

Simple Baseline has become a dominant benchmark in human pose estimation (HPE) due to its excellent performance and simple design. However, its “strong encoder + simple decoder” architectural paradigm suffers from two core limitations: (1) its non-branching, linear deconvolutional path prevents it from leveraging the rich, fine-grained features generated by the encoder at multiple scales and (2) the model lacks explicit prior knowledge of both the absolute positions and structural layout of human keypoints. To address these issues, this paper introduces AFJ-PoseNet, a new architecture that deeply enhances the Simple Baseline framework. First, we restructure Simple Baseline’s original linear decoder into a U-Net-like multi-scale fusion path, introducing intermediate features from the encoder via skip connections. For efficient fusion, we design a novel Attention Fusion Module (AFM), which dynamically gates the flow of incoming detailed features through a context-aware spatial attention mechanism. Second, we propose the Joint-Aware Positional Encoding (JAPE) module, which innovatively combines a fixed global coordinate system with learnable, joint-specific spatial priors. This design injects both absolute position awareness and statistical priors of the human body structure. Our ablation studies on the MPII dataset validate the effectiveness of each proposed enhancement, with our full model achieving a mean PCKh of 88.915, a 0.341 percentage point improvement over our re-implemented baseline. On the more challenging COCO val2017 dataset, our ResNet-50-based AFJ-PoseNet achieves an Average Precision (AP) of 72.6%. While this involves a slight trade-off in Average Recall for higher precision, this result represents a significant 2.2 percentage point improvement over our re-implemented baseline (70.4%) and also outperforms other strong, publicly available models like DARK (72.4%) and SimCC (72.1%) under comparable settings, demonstrating the superiority and competitiveness of our proposed enhancements. Full article

Heavy metal pollution, particularly copper contamination, threatens the ecological environment and human survival. In response to this pressing environmental issue, the development of innovative remediation strategies has become imperative. Bioremediation technology is characterized by remarkable advantages, including its ecological friendliness, cost-effectiveness, and operational efficiency. In our previous research, Rossellomorea sp. ZC255 demonstrated substantial potential for environmental bioremediation applications. This study investigated the removal characteristics and underlying mechanism of strain ZC255 and revealed that the maximum removal capacity was 253.4 mg/g biomass under the optimal conditions (pH 7.0, 28 °C, and 2% inoculum). The assessment of the biosorption process followed pseudo-second-order kinetics, while the adsorption isotherm may fit well with both the Langmuir and Freundlich models. Cell surface alterations on the Cu(II)-treated biomass were observed through scanning electron microscopy (SEM). Cu(II) binding functional groups were determined via Fourier transform infrared spectroscopy (FTIR) analysis. Simultaneously, the genomic analysis of strain ZC255 identified multiple genes potentially involved in heavy metal resistance, transport, and metabolic processes. These studies highlight the significance of strain ZC255 in the context of environmental heavy metal bioremediation research and provide a basis for using strain ZC255 as a copper removal biosorbent. Full article

Sexually transmitted infections (STIs) are a significant global health concern, affecting millions of people worldwide, particularly sexually active adolescents and young adults. These infections, caused by various pathogens, including bacteria, viruses, parasites, and fungi, can have profound implications for women’s reproductive health and fertility. This review explores the role of vaginal and uterine infections in women’s infertility, focusing on the most common pathogens and their impact on reproductive outcomes. Bacterial infections, such as those caused by intracellular bacteria (Mycoplasma, Ureaplasma, and Chlamydia), Neisseria gonorrhoeae, and bacterial vaginosis, are among the most prevalent causes of infertility in women. Studies have shown that these infections can lead to pelvic inflammatory disease, tubal occlusion, and endometrial damage, all of which can impair fertility. Mycobacterium tuberculosis, in particular, is a significant cause of genital tuberculosis and infertility in high-incidence countries. Viral infections, such as Human papillomavirus (HPV) and Herpes simplex virus (HSV), can also affect women’s fertility. While the exact role of HPV in female infertility remains unclear, studies suggest that it may increase the risk of endometrial implantation issues and miscarriage. HSV may be associated with unexplained infertility. Parasitic infections, such as trichomoniasis and schistosomiasis, can directly impact the female reproductive system, leading to infertility, ectopic pregnancy, and other complications. Fungal infections, such as candidiasis, are common but rarely have serious outcomes related to fertility. The vaginal microbiome plays a crucial role in maintaining reproductive health, and alterations in the microbial balance can increase susceptibility to STIs and infertility. Probiotics have been proposed as a potential therapeutic strategy to restore the vaginal ecosystem and improve fertility outcomes, although further research is needed to establish their efficacy. In conclusion, vaginal and uterine infections contribute significantly to women’s infertility, with various pathogens affecting the reproductive system through different mechanisms. Early diagnosis, appropriate treatment, and preventive measures are essential to mitigate the impact of these infections on women’s reproductive health and fertility. Full article

Due to human activities and the invasion of Spartina alterniflora, the population of Scirpus mariqueter (S. mariqueter) in the Yangtze River Estuary has gradually declined. To address this issue, numerous restoration efforts have been undertaken. To investigate the changes and influencing factors of soil bacterial communities during the restoration of S. mariqueter wetlands, we selected S. mariqueter populations as the research focus and divided the samples into two years, S1 and S2. High-throughput sequencing technology was employed for observation and analysis. The results revealed that from S1 to S2, soil bacterial diversity in the S. mariqueter wetland increased significantly and displayed clear seasonal patterns. The dominant bacterial phyla included Proteobacteria, Bacteroidota, Firmicutes, and Acidobacteriota. Among these, Proteobacteria had the highest relative abundance, while Acidobacteriota showed the most pronounced increase, surpassing Bacteroidota and Firmicutes to become the second most abundant group. Redundancy analysis (RDA) indicated that soil organic matter and electrical conductivity were the key factors influencing the composition and diversity of the soil bacterial community, with Acidobacteriota playing a dominant role during wetland restoration. In conclusion, during the ecological restoration of the S. mariqueter wetlands, attention should be given to environmental factors such as soil organic matter and electrical conductivity, while the regulatory role of Acidobacteriota in wetland soils should not be overlooked. This study provides a microscopic perspective on the interactions between microbial diversity and ecosystem functions in coastal wetlands, offering valuable guidance for the ecological restoration of degraded wetlands. Full article

Organoids allow to model healthy and diseased human tissues. and have applications in developmental biology, drug discovery, and cell therapy. Traditionally cultured in immersion/suspension, organoids face issues like lack of standardization, fusion, hypoxia-induced necrosis, continuous agitation, and high media volume requirements. To address these issues, we developed an air–liquid interface (ALi) technology for culturing organoids, termed AirLiwell. It uses non-adhesive microwells for generating and maintaining individualized organoids on an air–liquid interface. This method ensures high standardization, prevents organoid fusion, eliminates the need for agitation, simplifies media changes, reduces media volume, and is compatible with Good Manufacturing Practices. We compared the ALi method to standard immersion culture for midbrain organoids, detailing the process from human pluripotent stem cell (hPSC) culture to organoid maturation and analysis. Air–liquid interface organoids (3D-ALi) showed optimized size and shape standardization. RNA sequencing and immunostaining confirmed neural/dopaminergic specification. Single-cell RNA sequencing revealed that immersion organoids (3D-i) contained 16% fibroblast-like, 23% myeloid-like, and 61% neural cells (49% neurons), whereas 3D-ALi organoids comprised 99% neural cells (86% neurons). Functionally, 3D-ALi organoids showed a striking electrophysiological synchronization, unlike the heterogeneous activity of 3D-i organoids. This standardized organoid platform improves reproducibility and scalability, demonstrated here with midbrain organoids. The use of midbrain organoids is particularly relevant for neuroscience and neurodegenerative diseases, such as Parkinson’s disease, due to their high incidence, opening new perspectives in disease modeling and cell therapy. In addition to hPSC-derived organoids, the method’s versatility extends to cancer organoids and 3D cultures from primary human cells. Full article

Smart safety helmets equipped with vision systems are emerging as powerful tools for industrial infrastructure inspection. This paper presents a comprehensive state-of-the-art review of such VSLAM-enabled (Visual Simultaneous Localization and Mapping) helmets. We surveyed the evolution from basic helmet cameras to intelligent, sensor-fused inspection platforms, highlighting how modern helmets leverage real-time visual SLAM algorithms to map environments and assist inspectors. A systematic literature search was conducted targeting high-impact journals, patents, and industry reports. We classify helmet-integrated camera systems into monocular, stereo, and omnidirectional types and compare their capabilities for infrastructure inspection. We examine core VSLAM algorithms (feature-based, direct, hybrid, and deep-learning-enhanced) and discuss their adaptation to wearable platforms. Multi-sensor fusion approaches integrating inertial, LiDAR, and GNSS data are reviewed, along with edge/cloud processing architectures enabling real-time performance. This paper compiles numerous industrial use cases, from bridges and tunnels to plants and power facilities, demonstrating significant improvements in inspection efficiency, data quality, and worker safety. Key challenges are analyzed, including technical hurdles (battery life, processing limits, and harsh environments), human factors (ergonomics, training, and cognitive load), and regulatory issues (safety certification and data privacy). We also identify emerging trends, such as semantic SLAM, AI-driven defect recognition, hardware miniaturization, and collaborative multi-helmet systems. This review finds that VSLAM-equipped smart helmets offer a transformative approach to infrastructure inspection, enabling real-time mapping, augmented awareness, and safer workflows. We conclude by highlighting current research gaps, notably in standardizing systems and integrating with asset management, and provide recommendations for industry adoption and future research directions. Full article

This study develops and validates an artificial intelligence (AI)-assisted internship learning platform for automotive electronics based on the Llama 3 large language model, aiming to enhance pedagogical effectiveness within vocational training contexts. Addressing critical issues such as the persistent theory–practice gap and limited innovation capability prevalent in existing curricula, we leverage the natural language processing (NLP) capabilities of Llama 3 through fine-tuning based on transfer learning to establish a specialized knowledge base encompassing fundamental circuit principles and fault diagnosis protocols. The implementation employs the Hugging Face Transformers library with optimized hyperparameters, including a learning rate of 5 × 10⁻⁵ across five training epochs. Post-training evaluations revealed an accuracy of 89.7% on validation tasks (representing a 12.4% improvement over the baseline model), a semantic comprehension precision of 92.3% in technical question-and-answer assessments, a mathematical computation accuracy of 78.4% (highlighting this as a current limitation), and a latency of 6.3 s under peak operational workloads (indicating a system bottleneck). Although direct trials involving students were deliberately avoided, the platform’s technical feasibility was validated through multidimensional benchmarking against established models (BERT-base and GPT-2), confirming superior domain adaptability (F1 = 0.87) and enhanced error tolerance (σ² = 1.2). Notable limitations emerged in numerical reasoning tasks (Cohen’s d = 1.15 compared to human experts) and in real-time responsiveness deterioration when exceeding 50 concurrent users. The study concludes that Llama 3 demonstrates considerable promise for automotive electronics skills development. Proposed future enhancements include integrating symbolic AI modules to improve computational reliability, implementing Kubernetes-based load balancing to ensure latency below 2 s at scale, and conducting longitudinal pedagogical validation studies with trainees. This research provides a robust technical foundation for AI-driven vocational education, especially suited to mechatronics fields that require close integration between theoretical knowledge and practical troubleshooting skills. Full article

This paper addressed the problem of oil diagnostics lubricants applied to the predictive maintenance of industrial gearboxes, proposing the integration of an artificial intelligence (AI) system into the process analysis. The main objective was to overcome the critical issues of the traditional method, characterized by long analysis times and a marked dependence on the subjective interpretation of operators. The method includes a detailed statistical analysis of the common ways to assess the condition of lubricants, such as optical emission spectroscopy, particle counting, measuring viscosity and density, and Fourier-transform infrared spectroscopy (FT-IR). These methods are then combined with an artificial intelligence model. Tested on commercial gearbox data, the proposed approach demonstrates agreement between IA and expert evaluation. The application has shown that it can effectively support diagnoses, reduce processing time by 60%, and minimize human errors. It also improves knowledge sharing through an increase in the stability and repetitiveness of diagnoses and promotes consistency and clarity in reporting. Full article

Facing the demands of Human–Robot Collaborative (HRC) assembly for complex products under Industry 5.0, this paper proposes an intelligent assembly method that integrates Large Language Model (LLM) reasoning with Augmented Reality (AR) interaction. To address issues such as poor visibility, difficulty in knowledge acquisition, and strong decision dependency in the assembly of complex aerospace products within confined spaces, an assembly task model and structured process information are constructed. Combined with a retrieval-augmented generation mechanism, the method realizes knowledge reasoning and optimization suggestion generation. An improved ORB-SLAM2 algorithm is applied to achieve virtual–real mapping and component tracking, further supporting the development of an enhanced visual interaction system. The proposed approach is validated through a typical aerospace electronic cabin assembly task, demonstrating significant improvements in assembly efficiency, quality, and human–robot interaction experience, thus providing effective support for intelligent HRC assembly. Full article

Degradable fluorescent sensors present a promising portable approach for heavy metal ion detection, aiming to prevent secondary environmental pollution. Additionally, the excessive intake of ferric ions (Fe³⁺), an essential trace element for human health, poses critical health risks that urgently require effective monitoring. In this study, we developed a thermally degradable fluorescent hydrogel sensor (Eu-CDs@DPPG) based on europium-doped carbon dots (Eu-CDs). The Eu-CDs, synthesized via a hydrothermal method, exhibited selective fluorescence quenching by Fe³⁺ through the inner filter effect (IFE). Embedding Eu-CDs into the hydrogel significantly enhanced their stability and dispersibility in aqueous environments, effectively resolving issues related to aggregation and matrix interference in traditional sensing methods. The developed sensor demonstrated a broad linear detection range (0–2.5 µM), an extremely low detection limit (1.25 nM), and rapid response (<40 s). Furthermore, a smartphone-assisted LAB color analysis allowed portable, visual quantification of Fe³⁺ with a practical LOD of 6.588 nM. Importantly, the hydrogel was thermally degradable at 80 °C, thus minimizing environmental impact. The sensor’s practical applicability was validated by accurately detecting Fe³⁺ in spinach and human urine samples, achieving recoveries of 98.7–108.0% with low relative standard deviations. This work provides an efficient, portable, and sustainable sensing platform that overcomes the limitations inherent in conventional analytical methods. Full article

Behavioral issues in domestic dogs represent a significant welfare concern affecting both canines and their caregivers, with prevalence rates reported to range from 34 to 86% across the population. Current treatment options, including selective serotonin reuptake inhibitors (SSRIs) like fluoxetine, often present limitations including adverse effects and delayed efficacy. This randomized, placebo-controlled (maltodextrin) study investigated the effects of a novel Lactiplantibacillus plantarum strain (LP815^TM) on canine behavioral concerns through gut–brain axis modulation. Home-based dogs (n = 40) received either LP815^TM (n = 28) or placebo (n = 12) daily for 4 weeks, with behavioral changes assessed using the comprehensive Canine Behavioral Assessment & Research Questionnaire (C-BARQ) and continuous activity monitoring. After the intervention period, dogs receiving LP815^TM showed significant improvements in aggression (p = 0.0047) and anxiety (p = 0.0005) compared to placebo controls. These findings were corroborated by objective activity data, which demonstrated faster post-departure settling, reduced daytime sleep, and improved sleep consistency in the treatment group. Throughout >1120 administered doses, no significant adverse events were reported, contrasting favorably with pharmaceutical alternatives. The concordance between our findings and previous research using different L. plantarum strains suggests a consistent biological mechanism, potentially involving GABA production and vagal nerve stimulation. These results indicate that LP815^TM represents a promising, safe alternative for addressing common canine behavioral concerns with potential implications for improving both canine welfare and the human–animal bond. Full article

Cancer is a major global health issue, with rising incidence rates highlighting the urgent need for more effective treatments. Despite advances in cancer therapy, challenges such as adverse effects and limitations of existing treatments remain. Immunotherapy, which harnesses the body’s immune system to target cancer cells, offers promising solutions. Gamma delta (γδ) T cells are noteworthy due to their potent ability to kill various cancer cells without needing conventional antigen presentation. Recent studies have focused on the role of γδ T cells in α-galactosylceramide (α-GalCer)-mediated immunity, opening new possibilities for cancer immunotherapy. We engineered humanized T cell receptor (HuTCR)-T1 γδ mice by replacing mouse sequences with human counterparts. This study investigates the cytotoxic activity of humanized γδ T cells against several human cancer cell lines (A431, HT-29, K562, and Daudi) in vitro, aiming to elucidate mechanisms underlying their anticancer efficacy. Human cancer cells were co-cultured with humanized γδ T cells, with and without α-GalCer, for 24 h. The humanized γδ T cells showed enhanced cytotoxicity across all tested cancer cell lines compared to wild-type γδ T cells. Additionally, γδ T cells from HuTCR-T1 mice exhibited higher levels of anticancer cytokines (IFN-γ, TNF-α, and IL-17) and Granzyme B, indicating their potential as potent mediators of anticancer immune responses. Blocking γδ T cells’ cytotoxicity confirmed their γδ-mediated function. These findings represent a significant step in preclinical development of γδ T cell-based cancer immunotherapies, providing insights into their mechanisms of action, optimization of therapeutic strategies, and identification of predictive biomarkers for clinical application. Full article

Direct seeding has gained prominence as a labor-efficient and environmentally sustainable alternative to conventional transplanting in rice cultivation. In direct seeding systems, early-stage management is crucial for stable seedling establishment, with sowing uniformity measured by seed counts being a critical indicator of success. However, conventional manual seed counting methods are time-consuming, prone to human error, and impractical for large-scale or repetitive tasks, necessitating advanced automated solutions. Recent advances in computer vision technologies and precision agriculture tools, offer the potential to automate seed counting tasks. Nevertheless, challenges such as domain discrepancies and limited labeled data restrict robust real-world deployment. To address these issues, we propose a density estimation-based seed counting framework integrating semi-supervised learning and background augmentation. This framework includes a cost-effective data acquisition system enabling diverse domain data collection through indoor background augmentation, combined with semi-supervised learning to utilize augmented data effectively while minimizing labeling costs. The experimental results on field data from unknown domains show that our approach reduces seed counting errors by up to 58.5% compared to conventional methods, highlighting its potential as a scalable and effective solution for agricultural applications in real-world environments. Full article

Resource-exhausted cities have long played a key role in national energy development. Urban renewal projects, such as the renovation of old underground commercial spaces, can improve urban vitality and promote sustainable development. However, in resource-based cities, traditional industries dominate, while new industries such as modern commerce develop slowly. This results in low economic dynamism and weak motivation for urban development. To address this issue, we propose a systematic method for analyzing construction risks during the decision-making stage of renovation projects. The method includes three steps: risk value assessment, risk factor identification, and risk weight calculation. First, unlike previous studies that only used SWOT for risk factor analysis, we also applied it for project value assessment. Then, using the Work Breakdown Structure–Risk Breakdown Structure framework method (WBS-RBS), we identified specific risk sources by analyzing key construction technologies throughout the entire lifecycle of the renovation project. Finally, to enhance expert consensus, we proposed an improved Delphi–Analytic Hierarchy Process method (Delphi–AHP) to calculate risk indicator weights for different construction phases. The risk analysis covered all lifecycle stages of the renovation and upgrading project. The results show that in the Fushun city renovation case study, the established framework—consisting of five first-level indicators and twenty s-level indicators—enables analysis of renovation projects. Among these, management factors and human factors were identified as the most critical, with weights of 0.3608 and 0.2017, respectively. The proposed method provides a structured approach to evaluating renovation risks, taking into account the specific characteristics of construction work. This can serve as a useful reference for ensuring safe and efficient implementation of underground commercial space renovation projects in resource-exhausted cities. Full article

Microbiome analysis has revolutionized our understanding of various biological processes, spanning human health and epidemiology (including antimicrobial resistance and horizontal gene transfer), as well as environmental and agricultural studies. At the heart of microbiome analysis lies the characterization of microbial communities through the quantification of microbial taxa and their dynamics. In the study of bacterial abundances, it is becoming more relevant to consider their relationship, to embed these data in the framework of network theory, allowing characterization of features like node relevance, pathways, and community structure. In this study, we address the primary biases encountered in reconstructing networks through correlation measures, particularly in light of the compositional nature of the data, within-sample diversity, and the presence of a high number of unobserved species. These factors can lead to inaccurate correlation estimates. To tackle these challenges, we employ simulated data to demonstrate how many of these issues can be mitigated by applying typical transformations designed for compositional data. These transformations enable the use of straightforward measures like Pearson’s correlation to correctly identify positive and negative relationships among relative abundances, especially in high-dimensional data, without having any need for further corrections. However, some challenges persist, such as addressing data sparsity, as neglecting this aspect can result in an underestimation of negative correlations. Full article