Advancing Open Science

Youth soccer requires an integrated approach combining technical–tactical, physical, and psychological components to enhance performance and long-term engagement. Although Repeated Linear Sprint Training (LRST) and Repeated Change of Direction Speed (RCOD) training are widely used to improve fitness, direct comparisons of their effects on physical performance and perceptual responses in adolescent players remain limited. This study compared the effects of an 8-week LRST versus RCOD training program on physical performance, perceived exertion, and enjoyment in youth soccer players. Twenty-six male players were randomly assigned to an LRST group (n = 13) or an RCOD group (n = 13). Both groups completed two weekly sessions of their assigned training in addition to regular soccer practice. Pre- and post-intervention assessments included acceleration and sprint speed, change-of-direction (COD) performance (T-Half Test [THT], Illinois Agility Test [IAT]), lower-limb power (Five-Jump Test [5JT], Squat Jump [SJ], Countermovement Jump [CMJ]), and endurance-intensive fitness. Enjoyment and session-RPE were recorded after each training session. Both groups improved across all physical measures (main effect of time, p < 0.0001). Significant time × group interactions favored RCOD for THT (~1.6%), IAT (~1.1%), 5JT (~2.3%), CMJ (~5.2%), and SJ (~6.3%), with no overall main effect of group. Enjoyment was consistently higher in the RCOD group (p < 0.0001), while session-RPE did not differ between groups. In youth soccer, both LRST and RCOD effectively enhance physical performance. However, RCOD appears more effective for improving pre-planned COD and explosive performance while eliciting greater enjoyment without increasing perceived exertion. Incorporating structured RCOD training alongside linear sprint work may represent a practical strategy to optimize physical development and sustain player engagement.

Background/Objectives: Antimicrobial resistance (AMR) among Enterobacterales poses a major threat to public health in Southern Africa and has led to limited treatment options and increased mortality. Despite Africa bearing the brunt, there is limited data on the epidemiology and molecular epidemiology of the genetic determinants of β-lactam and/or carbapenem resistance. This narrative literature review summarizes the epidemiology and molecular characteristics of extended-spectrum β-lactamase-producing Enterobacterales (ESBL-PE), carbapenem-resistant Enterobacterales (CRE), and carbapenemase-producing Enterobacterales (CPE) in Southern Africa, while identifying data gaps and surveillance challenges. Methods: A comprehensive literature review was conducted using peer-reviewed articles from ten Southern African countries, including South Africa, Lesotho, Eswatini, Botswana, Namibia, Angola, Zambia, Zimbabwe, Mozambique, and Malawi, reporting the epidemiology and/or molecular characterization of ESBL-PE, CRE, and CPE. Results: ESBL-PE, CRE, and CPE pose an increasing healthcare threat in Southern Africa, with prevalence varying widely by source. Klebsiella pneumoniae and E. coli are the predominant ESBL-PE, CRE, and CPE species. The most frequent resistance genes are bla_CTX-M among ESBLs and bla_NDM and bla_OXA among carbapenemases, reflecting global patterns. However, molecular characterization across the region remains limited, with countries such as Botswana, Lesotho, Eswatini, Zambia, and Zimbabwe lacking sufficient data on the prevalence and diversity of these resistance determinants. Conclusions: Despite the paucity of genomic and epidemiological data, Southern Africa faces an urgent AMR challenge. Strengthening laboratory infrastructure, genomic surveillance, and regional coordination is crucial to mitigate AMR and guide antibiotic stewardship policies.

Objective: To evaluate the diagnostic utility of antepartum amnioinfusion in cases of severe oligo- and anhydramnios and to descriptively report perinatal outcomes. Methods: This retrospective single-center study analyzed all antepartum amnioinfusions performed between 2009 and 2024 in pregnancies between 16 + 0 and 34 + 0 weeks of gestation. The primary endpoint was diagnostic impact following amnioinfusion. Secondary endpoints were descriptive perinatal outcomes. No inferential statistical analyses were performed. Results: A total of 37 amnioinfusions were performed in 31 patients. Median gestational age at first amnioinfusion was 22 ± 4.3 weeks, with a mean infusion volume of 259 ± 59.4 mL. The most frequent etiologies were preterm prelabor rupture of membranes (PROM, 29%), fetal growth restriction (FGR, 25.8%), and urogenital malformations (22.6%). Amnioinfusion improved sonographic visualization and diagnostic assessment in the majority of cases. Pregnancy prolongation was observed in selected subgroups; however, causal inference regarding therapeutic efficacy cannot be drawn. Conclusions: Antepartum amnioinfusion represents a valuable adjunct for prenatal diagnostic evaluation in severe oligo- and anhydramnios. Observed perinatal outcomes should be interpreted descriptively. Further prospective, controlled studies are required to define the role of amnioinfusion beyond diagnostic feasibility.

Foodborne pathogens remain a major public health concern, and the early identification of unsafe conditions is essential for preventive control. Routine inspections generate rich textual and structured data that can support real-time assessment of pathogen-related risk. The objective of this study is to develop an explainable transformer-based framework for predicting food safety inspection grades using multimodal inspection data. We combine structured metadata with unstructured deficiency narratives and evaluate classical machine learning models, deep learning architectures, and transformer models. RoBERTa achieved the highest performance (F1 = 0.96), followed by BiLSTM (F1 = 0.95) and LightGBM (F1 = 0.92). SHapley Additive exPlanations (SHAP) analysis revealed linguistically meaningful indicators of pathogen-related hazards such as temperature abuse, pests, and unsanitary practices. The findings demonstrate that transformer-based models, combined with explainable AI (XAI), can support pathogen-oriented monitoring and real-time risk assessment. This study highlights the potential of multimodal AI approaches to enhance inspection efficiency and strengthen public health surveillance.

Influenza A virus (IAV) continues to present a threat to public health, highlighting the need for safe and multi-target antivirals. In this study, anti-influenza activity, airborne transmission blocking capacity, and immunomodulatory effects of Cnidium monnieri polysaccharides (CMP) were evaluated. Cytotoxicity in A549 cells was assessed by CCK-8 (CC₅₀ = 8.49 mg/mL), antiviral efficacy against A/California/04/2009 (CA04) by dose–response (EC₅₀ = 1.63 mg/mL), and the stage of action by time-of-addition assays (pre-, co-, post-treatment). A guinea pig model infected with CA04 was used for testing the effect of pre-exposure CMP on transmission, with readouts including nasal-wash titers, seroconversion, lung index, and tissue titers (EID₅₀). RT-qPCR was employed to quantify the mRNA expression levels of proinflammatory cytokines, including TNF-α, IL-1β, and IL-6, in lung tissue, while Western blot analysis was performed to assess the expression and phosphorylation status of key proteins involved in the NF-κB signaling pathway. CMP suppressed viral replication in vitro within non-cytotoxic ranges, and pre-treatment—rather than co- or post-treatment—significantly reduced titers and cytopathic effect, consistent with effects at pre-entry steps and/or host priming. In vivo, pre-exposure CMP lowered nasal shedding, reduced aerosol transmission (3/6 seroconverted vs. 6/6 controls), decreased lung indices, and diminished tissue viral loads; IAV was undetectable in trachea at 7 days post-infection in pre-exposed animals, and nasal-turbinate titers declined relative to infection controls. Moreover, during in vivo treatment in mice, CMP significantly suppressed the levels of inflammatory cytokines (TNF-α, IL-1β, and IL-6) in lung tissue. This effect was mechanistically associated with CMP-mediated regulation of the NF-κB signaling pathway, leading to attenuation of inflammatory responses. These data indicate that CMP combines a favorable in vitro safety and efficacy profile with inhibition of airborne spread in vivo, supporting further mechanistic, pharmacokinetic, and fractionation studies toward translational development.

Background: Computer-guided static implant surgery (CGSIS) is widely adopted to enhance the precision of dental implant placement. However, significant heterogeneity in reported accuracy values complicates evidence-based clinical decision-making. This variance is likely attributable to a fundamental lack of standardization in the methodologies used to assess dimensional accuracy. Objective: This scoping review aimed to systematically map, synthesize, and analyze the clinical methodologies used to quantify the dimensional accuracy of CGSIS. Methods: The review was conducted in accordance with the PRISMA-ScR guidelines. A systematic search of PubMed/MEDLINE, Scopus, and Embase was performed from inception to October 2025. Clinical studies quantitatively comparing planned versus achieved implant positions in human patients were included. Data were charted on study design, guide support type, data acquisition methods, reference systems for superimposition, measurement software, and accuracy metrics. Results: The analysis of 21 included studies revealed extensive methodological heterogeneity. Key findings included the predominant use of two distinct reference systems: post-operative CBCT (n = 12) and intraoral scanning with scan bodies (n = 6). A variety of proprietary and third-party software packages (e.g., coDiagnostiX, Geomagic, Mimics) were employed for superimposition, utilizing different alignment algorithms. Critically, this heterogeneity in measurement approach directly manifests in widely varying reported values for core accuracy metrics. In addition, the definitions and reporting of core accuracy metrics—specifically global coronal deviation (range of reported means: 0.55–1.70 mm), global apical deviation (0.76–2.50 mm), and angular deviation (2.11–7.14°)—were inconsistent. For example, these metrics were also reported using different statistical summaries (e.g., means with standard deviations or medians with interquartile ranges). Conclusions: The comparability and synthesis of evidence on CGSIS accuracy are significantly limited by non-standardized measurement approaches. The reported ranges of deviation values are a direct consequence of this methodological heterogeneity, not a comparison of implant system performance. Our findings highlight an urgent need for a consensus-based minimum reporting standard for future clinical research in this field to ensure reliable and translatable evidence.

The deployment of wireless seismic nodal systems necessitates the efficient identification of optimal locations for sensor installation, considering factors such as ground stability and the absence of interference. Semantic segmentation of satellite imagery has advanced significantly, and its application to this specific task remains unexplored. This work presents a baseline empirical evaluation of the U-Net architecture for the semantic segmentation of surfaces applicable for seismic sensor installation. We utilize a novel dataset of Sentinel-2 multispectral images, specifically labeled for this purpose. The study investigates the impact of pretrained encoders (EfficientNetB2, Cross-Stage Partial Darknet53—CSPDarknet53, and Multi-Axis Vision Transformer—MAxViT), different combinations of Sentinel-2 spectral bands (Red, Green, Blue (RGB), RGB+Near Infrared (NIR), 10-bands with 10 and 20 m/pix spatial resolution, full 13-band), and a technique for improving small object segmentation by modifying the input convolutional layer stride. Experimental results demonstrate that the CSPDarknet53 encoder generally outperforms the others (IoU = 0.534, Precision = 0.716, Recall = 0.635). The combination of RGB and Near-Infrared bands (10 m/pixel resolution) yielded the most robust performance across most configurations. Reducing the input stride from 2 to 1 proved beneficial for segmenting small linear objects like roads. The findings establish a baseline for this novel task and provide practical insights for optimizing deep learning models in the context of automated seismic nodal network installation planning.

The Calakmul Biosphere Reserve, in southeastern Mexico, is a major conservation area known for its tropical forests, emblematic wildlife species, and long history of Maya occupation. Established in 1989 as a federal Natural Protected Area, it was incorporated into UNESCO’s Man and the Biosphere Program in 1993 and designated a mixed World Heritage Site in 2014. Its socioecological trajectory is distinctive: conservation efforts advanced alongside the contemporary rural settlement resulting from agrarian reform and subsequent development and welfare policies. This article examines the persistent imbalance between ecological conservation and socioeconomic development surrounding the Calakmul Biosphere Reserve, focusing on water vulnerability in adjacent communities. The study integrates environmental history with household-level survey data on water access and vulnerability among 200 households in eight communities in the Biosphere Reserve’s transition zone, complemented by interviews with key water-management stakeholders. We document the consolidation of conservation through management plans, advisory councils, payments for ecosystem services, scientific research, and expanding voluntary conservation areas. Yet these advances contrast sharply with everyday socioeconomic realities: 68% of households face prolonged water scarcity, with an average of more than 30 days annually without water. Calakmul’s case highlights structural mismatch between conservation and local human well-being in Natural Protected Areas contexts.