With the widespread application of the engineering–procurement–construction (EPC) delivery model in large-scale infrastructure and complex industrial projects, the highly dynamic construction environment and the escalating complexity of risks pose heightened requirements for the competency of project safety directors. Conventional approaches that rely primarily on experience and manual inspections have become a bottleneck to further improvements in safety management capability. This paper investigates strategies to enhance the competency of safety directors in EPC projects through AI empowerment. Drawing on dynamic capability theory, synergy theory, and the Technology Acceptance Model (TAM), we develop a competency enhancement framework comprising three dimensions—sensing, seizing, and reconfiguring. A comparative experimental design and the Analytic Hierarchy Process (AHP) were employed for empirical evaluation. The results indicated that the Comprehensive Competency Index (CCI) under the AI-empowered mode increased by 45.9% relative to the traditional mode, enabling a shift in safety management from experience-driven practices to data- and intelligence-driven governance. Furthermore, this study proposes a tiered improvement strategy and a quantitative evaluation system, offering theoretical grounding and practical guidance for talent selection and digital-intelligent transformation in construction enterprises..

IntroductionAlzheimer’s disease (AD) prevalence rises with societal ageing. In clinical care, behavioral and psychological symptoms of dementia (BPSD)—including depression, agitation/aggression, apathy, and sleep disturbance—worsen patients’ quality of life and substantially increase caregiver burden, more significantly than the cognitive symptoms. Standard BPSD assessments rely on caregiver-rated questionnaires that are cross-sectional and may be biased when caregivers are themselves older adults. Device-based measures (e.g., research-grade wrist actigraphy) can provide objective longitudinal data and novel features. In parallel, therapeutic wearables may improve sleep and mood in adults, and might improve BPSD if accepted by people living with dementia.MethodsThis dyadic pilot study will recruit 20 participants (n = 10 AD patients; n = 10 caregivers) from outpatient services and affiliated day-care/dementia hubs in rural Taiwan. Participants will wear Geneactiv continuously for 8 weeks and Re-Timer ≥30 min/day for 4 weeks. Device-based data will be processed with GGIR, a well validated R-package designed for processing accelerometer data. Questionnaire assessments include Pittsburgh Sleep Quality Index (PSQI), Neuropsychiatry Inventory Questionnaire (NPI-Q), Caregiver Burden Inventory (CBI), and a semi-structured interview based on the Taiwanese version of Quebec User Evaluation of Satisfaction with Assistive Technology (T-QUEST) at prespecified timepoints.DiscussionWearable devices may facilitate the measurement and treatment of specific BPSD, as well as reduce caregiver burden. If proven feasible even in rural Taiwan where both digital and health literacy and resources are limited, this model will inform how device-based dementia care model can be considered and applied in the context of global ageing.

Implicit 3D geological modeling offers advantages in rapid model construction; however, purely data-driven approaches have limitations in modeling accuracy and often fail to represent complex geological structures faithfully. To enhance the accuracy and applicability of implicit 3D geological modeling, a method that integrates data-driven approaches with expert knowledge constraints is proposed. Using Leapfrog Geo as the platform, this study integrated multi-source data (boreholes, exploration line profiles) with geological expertise and a priori constraints, to achieve a high-precision reconstruction of the complex geological structure and 3D grade distribution in the Songshugang Ta-Nb deposit, Hengfeng, Jiangxi. Subsequently, deep-seated mineralization targets were predicted. The results demonstrate that this method significantly enhances the modeling accuracy and efficiency for intricate geological bodies. The constructed three-dimensional geological model indicates that the mineralization enrichment zones are located at the contact zones between sodic feldspar granite and the overlying potassic feldspar granite and greisenized granite, as well as between sodic feldspar granite and the underlying biotite granite. Based on this, two favorable mineral exploration targets have been predicted in the deeper regions.

This study investigates the coupled effects of loading rate and moisture condition on the mechanical behavior of cemented paste backfill (CPB). Uniaxial compression tests were conducted on CPB specimens under dry (0%), nature (18%), and saturated (32%) moisture conditions at loading rates ranging from 0.05 to 2.0 mm/min. The results show that uniaxial compressive strength (UCS) follows a second-order polynomial relation with loading rate and exhibits an overall rise-then-fall trend, with a common transition region around 1.0 mm/min. Relative to the nature condition, the dry condition generally enhances strength and stiffness, whereas the saturated condition weakens them but increases peak strain over the low-to-intermediate loading-rate range. Failure mode is also moisture dependent, with dry specimens showing more brittle behavior and saturated specimens exhibiting more tensile-dominated failure. The nonlinear index is likewise sensitive to loading rate and moisture condition, and saturated CPB exhibits the highest nonlinear index before the transition region. Based on UCPI analysis, the fitted optimal loading rates are 1.20, 1.23, and 0.97 mm/min for the dry, nature, and saturated conditions, respectively. Although the fitted optima differ slightly, they remain clustered around 1.0 mm/min. These findings indicate that loading-rate regulation in underground backfill-supported environments should account for the prevailing moisture condition rather than rely on a single fixed loading-rate criterion.

BackgroundThe quality of life of older adults is closely linked to functioning in the living environment. This study aims to assess the quality of life of older adults in southwestern Poland and the social factors determining it.MethodsThis cross-sectional study included a voluntary sample of 1108 older adults aged 60‒90. The WHOQoL-BREF questionnaire was used to assess quality of life.ResultsThe overall quality of life of the respondents (57%) was good. Men had a better quality of life in the psychological (69.0 ± 9.5 vs. 63.0 ± 6.5, p < 0.001) and environmental (69.0 ± 6.0 vs. 63.0 ± 9.5, p < 0.01) domains compared to women. In comparison, those living in a relationship had a better quality of life in the somatic (75.0 ± 9.0 vs. 69.0 ± 9.0, p = 0.034), social (75.0 ± 12.5 vs. 69.0 ± 9.5, p < 0.001), and environmental (69.0 ± 6.0 vs. 63.0 ± 9.5, p < 0.001) domains. Those with higher education had a better quality of life in the somatic (75.0 ± 6.0 vs. 69.0 ± 9.0 and 69.0 ± 9.0, p < 0.001), psychological (69.0 ± 6.5 vs. 63.0 ± 9.5 and 63.0 ± 6.5, p < 0.001), and environmental (69.0 ± 6.0 vs. 63.0 ± 9.5 and 63.0 ± 6.5, p < 0.001) domains compared to those with secondary and primary education. Respondents in good health had a better quality of life in all domains.ConclusionsGood quality of life in older adults surveyed is determined by male gender, marital status in a relationship, higher education, good health ratings, and fewer chronic diseases. Our results can guide policymakers, highlighting the fact that good health, fewer chronic diseases, and higher education, translate into a better quality of life. Therefore, there should be improved access to education for all people. And good population health should be prioritized by decision-makers.Trial registrationISRCTN platform as 18225729; January 2021.

Tea (Camellia sinensis) is the world’s second most consumed beverage, enjoyed daily by more than two billion people. In Bangladesh, it serves as a cornerstone agricultural export and a major sector of the domestic economy. However, commercial tea cultivation remains highly vulnerable to fungal and pest-related diseases such as Blight, Red Rust, and Helopeltis which severely reduce crop yield and compromise leaf quality. While early detection is critical to preventing widespread outbreaks, traditional manual inspection is slow, subjective, and highly error-prone. Deep learning provides a scalable alternative, yet single-branch networks often struggle to capture both minute disease lesions and broader structural degradation simultaneously. To address this, we propose a Hybrid Feature Fusion architecture that runs two highly efficient feature extractors in parallel: EfficientNetV2-Small to isolate fine-grained local textures, and MobileNetV3-Small to capture the global structural context of the leaf. The models were trained and evaluated on a real-world dataset of 2,000 annotated images, evenly distributed across the four target classes (Blight, Red Rust, Helopeltis, and Healthy). Before training, the images underwent a standardized preprocessing pipeline including resizing to 224 × 224 pixels and normalization, supplemented by a dynamic augmentation strategy featuring random rotations, horizontal flips, and brightness adjustments to improve model robustness. The proposed hybrid framework achieved an outstanding peak classification accuracy of 96.80% alongside a macro Area Under the Curve (AUC) of 0.9980. To rigorously validate its performance, the hybrid model was benchmarked against six diverse architectures: a Vision Transformer (ViT-B16 at 76.40%), a Custom CNN (89.60%), MobileNetV3 (94.40%), ResNet50 (95.60%), DenseNet121 (96.40%), and EfficientNetV2-B3 (97.60%). Although EfficientNetV2-B3 achieved a marginally higher raw accuracy, the proposed dual-branch framework delivered a superior precision-recall balance and faster convergence stability. These findings demonstrate that the proposed hybrid methodology is highly reliable and computationally balanced, making it an ideal candidate for integration into Internet of Things (IoT) edge devices for real-time disease monitoring in precision agriculture.

ObjectivesThis study explores the perspectives and experiences of healthcare professionals (HCPs) in a multidisciplinary fetal medicine (FM) team when delivering unexpected news in fetal medicine, with a particular focus on pregnant people living with mental health conditions (MHCs). It aims to identify communication strategies and highlight facilitators and barriers to effective consultations in these challenging situations.MethodsQualitative study involving 20 semi-structured interviews with HCPs from multi-disciplinary FM teams in a large NHS Trust in East London. Participants included FM specialists, neonatologists, midwives, and specialist nurses. Data were analyzed using reflexive thematic analysis to identify key themes representing participants’ perspectives.ResultsFive key themes emerged: (1) enhancing patient understanding, (2) supporting patients (3) individualizing care, (4) supporting HCPs and (5) the impact of health inequalities in FM. Facilitators of effective communication included optimizing the setting, avoiding ambiguity, fostering understanding, and involving significant others in discussions. Barriers included language limitations, where reliance on interpreters hindered effective and compassionate communication, and poor pregnancy-related health literacy, particularly in socio-economically deprived populations, which impeded informed decision-making. Clinician-related barriers included reluctance to address MHC due to lack of training, stigma around mental health, and time constraints. The emotional toll of frequently delivering distressing news was identified as a significant challenge for HCPs, yet formal support mechanisms were lacking.ConclusionHCPs in FM emphasized that delivering unexpected news requires a patient-centred approach that considers individual needs, is culturally sensitive, and addresses language and health literacy barriers. The mental health and well-being of both those receiving and delivering the news remains largely overlooked. Currently, no specific communication training exists for HCPs in FM and limited or no support is available for individuals with mild, pre-existing, or suspected MHC, despite their prevalence and the recognition that difficult news during pregnancy may exacerbate these conditions. Future research should evaluate communication frameworks co-produced by HCPs and individuals with lived experience to improve FM consultations for patients with and without MHC. Enhanced mental health support is urgently needed for both patients and HCPs in FM.

BackgroundGlioblastoma (GBM) is the most prevalent and malignant primary brain tumor in adults. While immune evasion is a well-recognized driver of GBM progression and a major obstacle for efficient immunotherapy, the role of the complement system remains underexplored. C1-inhibitor (C1-INH), a regulator of complement activation, was recently found overexpressed in GBM. We therefore hypothesized that GBM overexpresses additional complement regulators beyond C1-INH and the present work aimed to identify these.MethodGene expression of complement inhibitors, the complement regulator pentraxin-3 (PTX3), and complement proteins was analyzed across nine publicly available transcriptomic datasets. Within each dataset, statistical comparisons were performed between sample groups for each gene. Differentially expressed complement inhibitors were validated at the protein level by immunostaining in the rat GBM cell line NS1 and patient derived GBM tissue.ResultsCFI, encoding factor I, was significantly overexpressed in GBM compared to non-tumoral brain, while THBD and CFH, encoding thrombomodulin and factor H, displayed moderate overexpression. SERPING1, encoding C1-INH, was also upregulated, confirming previous findings. Immunostaining confirmed the expression of these inhibitors in vitro as well as in human glioblastoma tissue. Additionally, PTX3 and early complement proteins were significantly overexpressed in GBM, while levels of C5 and downstream components were comparable to normal brain.ConclusionsOur findings indicate that the GBM tumor overexpresses a specific set of complement regulators and components of the complement cascade, possibly inhibiting an efficient anti-tumoral immune response. Further investigations of these regulators as potential therapeutical targets in GBM are therefore highly warranted.

BackgroundAchieving and sustaining viral load suppression (VLS) among children living with HIV remains challenging despite the availability of antiretroviral therapy (ART), primarily due to adherence difficulties. This study evaluated the effects of three interventions under the ACHIEVE project on viral load outcomes (viral rebound and viral suppression) among HIV-positive orphaned and vulnerable children aged 0–17 years (hereafter referred to as CLHIV) receiving ART in Tanzania.MethodsThis is a longitudinal analysis of 26,257 CLHIV who are beneficiaries of the ACHIEVE project in Tanzania, each with at least two viral load (VL) test results between 2021 and 2023. VL outcomes were assessed using two analytic groups: Group 1 included 21,448 CLHIV with undetectable VL (<50 copies/mL) at baseline, where those maintaining undetectable VL were compared with those experiencing viral rebound (VL ≥ 50 copies/mL) at follow-up. Group 2 comprised 4,809 CLHIV with detectable VL (≥50 copies/mL) at baseline, where those achieving undetectable VL at follow-up were compared with those who did not. The study used a multivariable logistic regression model and propensity score matching (PSM) to evaluate the effects of ACHIEVE project interventions–WORTH Yetu economic strengthening, linkage to teen/paediatric clubs, and health insurance–on viral rebound (Group 1), and the achievement of an undetectable viral load (Group 2).FindingsOverall, 13.1% (2,805/21,448) of Group 1 CLHIV experienced viral rebound at follow-up. CLHIV who received ACHIEVE interventions were 23.2% less likely to experience viral rebound compared to those who did not (aOR = 0.768, 95% CI 0.668–0.882, p < 0.001). In Group 2, 70.9% (3,411/4,809) achieved an undetectable viral load at follow-up, with those who received ACHIEVE interventions being 31.9% more likely to achieve an undetectable viral load compared to those who did not (aOR = 1.319, 95% CI 1.059–1.643, p = 0.014). These findings were confirmed in the PSM analysis (viral rebound: β = – 0.047, 95% CI –0.072 – –0.021, p < 0.001; achieving undetectable viral load: β = 0.10, 95% CI 0.04–0.16, p = 0.001).ConclusionThe ACHIEVE project interventions were significantly associated with a reduced likelihood of viral rebound, and an increased likelihood of achieving undetectable viral load among CLHIV, highlighting their potential to enhance ART outcomes. These findings suggest that expanding similar community-based interventions could further contribute to HIV treatment efficacy and support more children in achieving and maintaining an undetectable viral load, particularly those at risk of viral rebound or persistent high viral load. Future research should explore strategies for scaling these interventions and evaluating their long-term impact.

BackgroundEnterotoxigenic Escherichia coli (ETEC) remains a major cause of diarrheal disease in low- and middle-income countries (LMICs). To curb ETEC related diarrhoea, several candidate vaccines are in development, with ETVAX® being the most advanced. Although immunogenicity studies have primarily focused on measuring antibody titres, assessing antibody avidity offers additional valuable insight into antibody quality and immune maturity. This study assessed anti-CS3 and anti-LTB IgG avidity in Zambian children to better understand vaccine-induced antibody responses in an endemic setting.MethodsChildren aged 6–23 months (n = 60) received three quarter-doses of ETVAX® with dmLT adjuvant on days 1, 14, and 90. Plasma samples collected at baseline prior to the first vaccination (Day 1; V1), seven days after the second dose (Day 22; V5), and seven days after the third dose (Day 97; V7) were analysed by limiting antigen dilution ELISA to calculate avidity indices (AI). Naïve classification was performed using titre-based thresholds (20th percentile of baseline titres) and avidity-defined naivety (AI < 0.5). Receiver operating characteristic (ROC) analysis was used to evaluate the discriminatory performance of avidity indices against titre-defined naïve status.ResultsBaseline avidity was detectable for both CS3 and LTB, consistent with prior natural exposure. Mean CS3 IgG avidity decreased from 0.7 at baseline to 0.6 after the third dose (p < 0.001), while LTB IgG avidity showed transient decreases but no net gain. Naïve classification at baseline revealed that 9/60 children had titres but low avidity (functional naivety), and 6/60 had waned titres but high avidity. Only one child was naïve by both criteria for CS3, and none for LTB. ROC analysis demonstrated moderate discrimination for CS3 (AUC = 0.65; optimal cut-off AI = 0.36) but poor discrimination for LTB (AUC = 0.30).ConclusionIn this endemic population, ETVAX® induced strong antibody titres but minimal changes in avidity over time, with notable inter-individual variability. ROC analysis highlighted context-specific limitations in using avidity to discriminate immune maturity. Together, these findings suggest that while antibody avidity may provide complementary information on antibody quality, its interpretation should be cautious and considered alongside antibody titres in endemic settings.