Search Results (283)

The existing deep learning-based Airborne Laser Scanning (ALS) point cloud semantic segmentation methods require a large amount of labeled data for training, which is not always feasible in practice. Insufficient training data may lead to over-fitting. To address this issue, we propose a novel Multi-branch Feature Extractor (MFE) and a three-stage transfer learning strategy that conducts pre-training on multi-source ALS data and transfers the model to another dataset with few samples, thereby improving the model’s generalization ability and reducing the need for manual annotation. The proposed MFE is based on a novel multi-branch architecture integrating Neighborhood Embedding Block (NEB) and Point Transformer Block (PTB); it aims to extract heterogeneous features (e.g., geometric features, reflectance features, and internal structural features) by leveraging the parameters contained in ALS point clouds. To address model transfer, a three-stage strategy was developed: (1) A pre-training subtask was employed to pre-train the proposed MFE if the source domain consisted of multi-source ALS data, overcoming parameter differences. (2) A domain adaptation subtask was employed to align cross-domain feature distributions between source and target domains. (3) An incremental learning subtask was proposed for continuous learning of novel categories in the target domain, avoiding catastrophic forgetting. Experiments conducted on the source domain consisted of DALES and Dublin datasets and the target domain consists of ISPRS benchmark dataset. The experimental results show that the proposed method achieved the highest OA of 85.5% and an average F1 score of 74.0% using only 10% training samples, which means the proposed framework can reduce manual annotation by 90% while keeping competitive classification accuracy. Full article

Background: Sports performance indicators are mainly based on male athletes, highlighting the importance of portraying the female reality, particularly in rural contexts. This study aims to characterize sports performance indicators (body composition and physical fitness) of young Portuguese female athletes. Methods: A cross-sectional study was conducted with 124 girls (13.66 ± 1.93 years) participating in federated team sports in a rural region of Portugal. Body composition was assessed using bioelectrical impedance, and physical fitness was evaluated through vertical jump tests (countermovement jump and squat jump), sprint (20 m), agility (T-test), handgrip strength, and cardiovascular endurance (Yo-Yo IR1). Results: Volleyball players are taller; football and basketball players are heavier; football and volleyball players have more fat-free mass than handball players (p < 0.05). Body mass index and % body fat did not differ between sports (p > 0.05). Volleyball players performed better in the countermovement jump (F = 4.146, p = 0.008) and squat jump (F = 7.686, p < 0.001) when compared to basketball, football, and handball players. No differences were observed in the speed or cardiorespiratory endurance tests (p > 0.05). Conclusions: The results revealed that, despite some specific differences between sports, most physical fitness indicators did not differ significantly between sports after controlling for age, menarche, and training experience. These findings suggest that shared contextual limitations in rural regions may take precedence over sport-specific adaptations in the early stages of sports participation. Full article

Objectives: Shoulder injury prevalence appears to be the highest among all injuries in CrossFit (CF) athletes. Nevertheless, there is no evidence deriving from prospective studies to explain this phenomenon. The purpose of this study was to document shoulder injury incidence in CF participants over a 12-month period and prospectively investigate the risk factors associated with their demographic, epidemiological, and functional characteristics. Methods: The sample comprised 109 CF athletes in various levels. Participants’ data were collected during the baseline assessment, using a specially designed questionnaire, as well as active range of motion, muscle strength, muscle endurance, and sport-specific tests. Non-parametric statistical tests and inferential statistics were employed, and in addition, linear and regression models were created. Logistic regression models incorporating the study’s continuous predictors to classify injury occurrence in CF athletes were developed and evaluated using the Area Under the ROC Curve (AUC) as the performance metric. Results: A shoulder injury incidence rate of 0.79 per 1000 training hours was recorded. Olympic weightlifting (45%) and gymnastics (35%) exercises were associated with shoulder injury occurrence. The most frequent injury concerned rotator cuff tendons (45%), including lesions and tendinopathies, exhibiting various severity levels. None of the examined variables individually showed a statistically significant correlation with shoulder injuries. Conclusions: This is the first study that has investigated prospectively shoulder injuries in CrossFit, creating a realistic profile of these athletes. Despite the broad spectrum of collected data, the traditional statistical approach failed to identify shoulder injury predictors. This indicates the necessity to explore this topic using more sophisticated techniques, such as advanced machine learning approaches. Full article

Background and Objectives: Excessive social media use is repeatedly linked to negative body image outcomes, yet its association with muscle dysmorphia, especially in athletic youth, remains underexplored. We investigated how social media exposure, comparison behavior, and platform engagement relate to muscle dysmorphia symptomatology in young German athletes. Materials and Methods: An anonymous, web-based cross-sectional survey was conducted (July–October 2024) of 540 individuals (45% female; mean age = 24.6 ± 5.3 years; 79% ≥ 3 h sport/week) recruited via Instagram. The questionnaire comprised demographics, sport type, detailed social media usage metrics, and the validated German Muscle Dysmorphic Disorder Inventory (MDDI-1, 15 items). Correlations (Spearman’s ρ, Kendall’s τ) were calculated; multivariate importance was probed with classification-and-regression trees and CatBoost gradient boosting, interpreted via SHAP values. Results: Median daily social media time was 76 min (IQR 55–110). Participants who spent ≥ 60 min per day on social media showed higher MDDI scores (mean 38 ± 7 vs. 35 ± 6; p = 0.010). The strongest bivariate link emerged between perceived social media-induced body dissatisfaction and felt pressure to attain a specific body composition (Spearman ρ = 0.748, Kendall τ = 0.672, p < 0.001). A CatBoost gradient-boosting model out-performed linear regression in predicting elevated MDDI. The three most influential features (via SHAP values) were daily social media time, frequency of comparison with fitness influencers, and frequency of “likes”-seeking behavior. Conclusions: Intensive social media exposure substantially heightens muscle dysmorphia risk in young German athletes. Machine-learning interpretation corroborates time on social media and influencer comparisons as primary drivers. Interventions should combine social media literacy training with sport-specific psychoeducation to mitigate maladaptive comparison cycles and prevent downstream eating disorder pathology. Longitudinal research is warranted to clarify causal pathways and to test targeted digital media interventions. Full article

Background/Objectives: Injury prevention programs (IPPs) are evidence-based interventions that reduce musculoskeletal injuries in youth sports. Despite their proven benefits, the adoption of IPPs by coaches remains limited. This study aimed to evaluate the awareness, usage, and perceptions of IPPs among youth sports coaches in Poland and to identify factors associated with their implementation. Methods: A cross-sectional study was conducted using a web-based survey tailored to youth sports coaches in Poland. Coaches of athletes aged 9–17 were recruited through targeted outreach to clubs and professional networks. The survey assessed IPP awareness, implementation, perceptions, and sources of information. Statistical analyses included chi-square tests, non-parametric comparisons, Firth’s logistic regression, and cluster profiling. Results: Only 54.6% of coaches (59 out of 108) were aware of IPPs, and among them, just 47.5% reported using them. No significant associations were found between IPP use and demographic variables such as gender, sport, or place of residence. Coaches who were aware of IPPs were significantly younger than those who were unaware (p = 0.029). The information source was the strongest predictor of IPP implementation: coaches trained via formal courses were over 20 times more likely to use IPPs compared to those learning from peers (OR = 20.4, p < 0.001). While coaches generally perceived IPPs as beneficial for fitness and recovery, 28.6% expressed doubts about their effectiveness in reducing injury risk. Conclusions: Despite broadly positive beliefs, only 47.5% of coaches who were aware of IPPs reported using them. Formal training significantly enhances the likelihood of adoption. These findings emphasize the need for structured educational efforts and improved dissemination strategies to promote evidence-based injury prevention in youth sports settings. Full article

Background: Dog agility is a rapidly growing sport involving a partnership between a dog and the handler, running through an obstacle course. Despite its increasing popularity and physical benefits, research on handler injuries remains limited. This study aimed to assess injury epidemiology of athletes practicing dog agility. Methods: This cross-sectional study was conducted using a comprehensive online survey consisting of 124 items, available in both English and Italian. The questionnaire was divided into four sections: Introduction collected demographic data and medical history; Materials and Methods focused on agility-related activities; Results explored injuries sustained in the past 12 months; Discussion examined training habits unrelated to agility. Results: Among 389 participants, the most represented age group ranged between 30 and 40 years old. Overall, 7% reported upper limb injuries, while 27% experienced at least one lower limb injury. Additionally, 20% of participants used medication, and 25% reported at least one chronic illness. On average, handlers trained twice per week and competed in two events per month. Lower limb injuries were predominantly muscular (49%) or ligamentous (14%) and most commonly occurred on grass pitches (56%). These injuries were more common in participants with a higher BMI, those using dynamic handling styles, and those competing at higher levels. Conclusions: This cross-sectional study highlighted the importance of identifying risk factors associated with dog agility handlers. Lower limb injuries were the most common, often associated with increased physical demands and handling styles involving intensive running and correlated with reduced physical fitness. Athletic conditioning, including structured warm-up and cool-down practices, might help decline injury risks. Full article

Introduction: Work life quality (WLQ) in nursing is a critical factor that influences both staff well-being and the quality of care provided to patients. Spiritual intelligence (SI) and servant leadership (SL) have been identified as potential positive predictors of WLQ, as they facilitate resilience, job satisfaction, and stress management in highly demanding hospital environments. However, the specific relationship between these constructs in the Peruvian nursing context has not yet been thoroughly explored. Objective: We aimed to examine the impact of spiritual intelligence and servant leadership on the work life quality of Peruvian nurses, assessing their predictive role through a structural equation modeling approach. Methods: A cross-sectional and explanatory study was conducted with a sample of 134 Peruvian nurses (M = 36.29 years, SD = 7.3). Validated Spanish-language instruments were used to measure SI, SL, and WLQ. Structural equation modeling (SEM) with a robust maximum likelihood estimator (MLR) was employed to evaluate the relationships between the variables. Results: Spiritual intelligence showed a positive correlation with WLQ (r = 0.40, p < 0.01) and with servant leadership (r = 0.44, p < 0.01). Likewise, servant leadership had a significant relationship with WLQ (r = 0.53, p < 0.01). The structural model demonstrated a good fit (χ² = 1314.240, df = 970, CFI = 0.96, TLI = 0.96, RMSEA = 0.05, SRMR = 0.08). The hypothesis that SI positively predicts WLQ was confirmed (β = 0.41, p < 0.001), as was the significant effect of SL on WLQ (β = 0.26, p < 0.001). Conclusions: The results indicate that both spiritual intelligence and servant leadership are key predictors of work life quality in Peruvian nurses. SI contributes to developing a transcendent perspective on work and greater resilience, while SL fosters a positive and motivating organizational environment. It is recommended to implement training programs and leadership strategies focused on these constructs to enhance work life quality in the healthcare sector. Full article

Objectives: The aim of this study was to validate the Polish version of the Somatosensory Amplification Scale (SSAS-PL) and examine its psychometric properties in clinical and non-clinical samples. Methods: The study included 1128 participants (711 healthy adults, 194 cardiac patients, 223 psychiatric patients). The analyses were categorized into exploratory and confirmatory phases. Exploratory analyses were conducted on a randomly selected sample that comprised 60% of the study participants (training sample) to estimate the reliability (Cronbach’s alpha) and factorial validity (EFA with varimax rotation). Confirmatory analyses were performed on an independent (test) sample that represented 40% of the total sample size to facilitate the cross-validation of the factor structure (CFA) and to assess the convergent and discriminant validities (using the HTMT method) in relation to health anxiety (SHAI) and psychopathological symptoms (KOFF-58). Additionally, measurement invariance was examined with respect to gender (female vs. male) and health status (healthy vs. clinical). Results: The SSAS-PL demonstrated good internal consistency (α = 0.75–0.78) after removing item 1. A one-factor structure showed the best fit and theoretical interpretability. The measurement invariance was supported across clinical groups. The SSAS-PL showed convergent validity with the measures of somatic symptoms, anxiety, and health anxiety. It demonstrated discriminant validity from other psychopathology measures. Conclusions: The SSAS-PL was a reliable and valid measure of somatosensory amplification in the Polish population. Its unidimensional structure aligned with most cross-cultural adaptations. The scale may be useful for assessing somatosensory amplification in both research and clinical settings in Poland. Further research on its utility in specific clinical populations is warranted. Full article

Background: Headache disorders, particularly migraine, are a leading cause of disability among active-duty military personnel, significantly affecting operational readiness and fitness for duty. Despite their high prevalence, limited data exist on how headache disorders are managed within military primary care systems. This study aimed to evaluate diagnostic confidence, treatment strategies, and structural challenges in the management of headache disorders from the perspective of military primary care physicians. Methods: A prospective, nationwide cross-sectional survey was conducted between May and July 2023 among all active-duty military physicians in primary care roles. An anonymous 15-item questionnaire assessed diagnostic practices, therapeutic approaches, referral pathways, perceived knowledge gaps, and suggestions for system improvements. The survey was distributed across military medical centers and outpatient clinics in Germany. Results: Ninety military physicians participated. Migraine and tension-type headache were commonly encountered, with 70% having treated at least one headache patient in the week prior to the survey. Diagnostic confidence was high for migraine (83.4%) and tension-type headache (77.8%) but lower for medication-overuse headache (65.5%) and cluster headache (47.8%). Acute treatment was widely implemented, but only 27.8% of respondents regularly initiated preventive therapies. Awareness of clinical guidelines was limited: only 23.3% were familiar with the ICHD-3, and just 58.9% with national headache treatment guidelines. Respondents expressed strong demand for targeted education, practical diagnostic tools, and improved interdisciplinary coordination. Conclusions: Headache disorders are a prevalent and clinically significant issue in military primary care. While military physicians show high engagement, important gaps exist in preventive treatment, guideline familiarity, and access to specialist care. Structured training, standardized treatment protocols, and system-level improvements are essential to optimize headache care and maintain operational readiness. Full article

The rising prevalence of elderly obesity in developed countries poses a public health challenge, since body composition changes during aging are associated with higher risks of chronic diseases. We cross-sectionally explored the relationship between diet, physical activity, and sex-specific differences in body composition among 378 elderly previously enrolled in the Florence European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Information on dietary habits and lifestyle was collected through validated questionnaires. Adherence to the Italian Mediterranean Index (IMI), Dietary Approaches to Stop Hypertension (DASH), and Greek Modified Mediterranean Diet (GMMD) a priori dietary patterns was calculated. Anthropometric measures were taken by trained personnel, and body composition parameters were estimated via bioelectrical impedance. In age- and energy-intake-adjusted regression models, adherence to the DASH and IMI patterns was associated with healthier body composition among women, while no significant relationship emerged among men. Fitness activities and total recreational physical activity revealed positive associations with healthier body composition (lower % fat mass, higher % muscle mass, and reduced waist circumference) in both sexes. These findings highlight the synergistic effect of diet and physical activity on body composition in the elderly and underscore the need for sex-specific interventions for promoting healthy aging. Full article

The short-term scheduling of pumped-storage hydropower plants is characterised by high dimensionality and nonlinearity and is subject to multiple operational constraints. This study proposes an intelligent scheduling framework that integrates an Atomic Orbital Search (AOS)-optimised Long Short-Term Memory (LSTM) network with the Deep Deterministic Policy Gradient (DDPG) algorithm to minimise water consumption during the generation period while satisfying constraints such as system load and safety states. Firstly, the AOS-LSTM model simultaneously optimises the number of hidden neurons, batch size, and training epochs to achieve high-precision fitting of unit flow–efficiency characteristic curves, reducing the fitting error by more than 65.35% compared with traditional methods. Subsequently, the high-precision fitted curves are embedded into a Markov decision process to guide DDPG in performing constraint-aware load scheduling. Under a typical daily load scenario, the proposed scheduling framework achieves fast inference decisions within 1 s, reducing water consumption by 0.85%, 1.78%, and 2.36% compared to standard DDPG, Particle Swarm Optimisation, and Dynamic Programming methods, respectively. In addition, only two vibration-zone operations and two vibration-zone crossings are recorded, representing a reduction of more than 90% compared with the above two traditional optimisation methods, significantly improving scheduling safety and operational stability. The results validate the proposed method’s economic efficiency and reliability in high-dimensional, multi-constraint pumped-storage scheduling problems and provide strong technical support for intelligent scheduling systems. Full article

This study validates a Spanish-language instrument designed to assess self-efficacy, digital competence, and anxiety in data-driven decision-making (DDDM) among pre-service teachers. Based on the 3D-MEA and the Beliefs about Basic ICT Competencies scale, the instrument was culturally adapted for Chile and Spain. A sample of 512 participants underwent exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Given the ordinal nature of the data and the assumption of non-normality, appropriate estimation methods were utilized. Results supported a well-defined four-factor structure: Interpretation and Application, Technology, Identification, and Anxiety. Factor loadings ranged from 0.678 to 0.869, and internal consistency was strong (α = 0.802–0.888). The CFA confirmed good model fit (χ² (129) = 189.25, p < 0.001; CFI = 0.985; TLI = 0.981; RMSEA = 0.041; SRMR = 0.061). Measurement invariance was established across gender and nationality, reinforcing the validity of cross-group comparisons. The study is framed within an educational context aligned with socioformative principles and sustainable education goals, which support reflective and ethical data use. This validated tool addresses the lack of culturally adapted and psychometrically validated instruments for assessing DDDM competencies in Spanish-speaking contexts, offering a culturally and linguistically relevant instrument with strong internal consistency and a well-supported factor structure. It supports the design of formative strategies in teacher education, enabling the identification of training needs and promoting evidence-based pedagogical decision-making in diverse Hispanic contexts. Future studies should test factorial invariance across additional contexts and explore longitudinal applications. Full article

Vitamin D is a key micronutrient in the function of the skeletomuscular system. Athletes are at increased risk of developing vitamin D deficiency during the execution of very demanding disciplines such as CrossFit^®. Single-nucleotide polymorphisms (SNPs) may influence circulating 25-hydroxy-vitamin D (25(OH)D) levels. An observational, longitudinal pilot study was conducted with 50 trained males according to specific inclusion criteria. Blood samples were obtained to determine 25(OH)D, vitamin D-binding protein (VDBP), vitamin D-receptor (VDR)circulating levels, and the presence of SNPs after DNA isolation and genotyping: rs10741657 to CYP2R1, rs2282679 to GC and rs2228570 to VDR genes. Significant differences (p < 0.05) in 25(OH)D concentration were determined between the biallelic combinations of rs228679 (GC) and rs228570 (VDR). The VDBP and VDR proteins did not show different levels in the case of the rs10741657 (CYP2R1) alleles. Statistically significant weak positive correlations (p < 0.05) were observed between 25(OH)D and AA-alleles of the CYP2R1 and VDR genes, and TT-alleles of the GC gene. Additionally, AA (rs10741657 and rs2228570) and TT (rs2282679) have a probability between 2 and 4 of having major effects on the concentration of 25(OH)D. Conversely, GG alleles present a probability of suboptimal values of 25(OH)D of 69%, 34%, and 24% for VDR, GC, and CYP2R1, respectively, showing a strong moderate positive correlation (r = 0.41) between the degrees of sports performance and 25(OH)D plasma levels. CYP2R1 (rs10741657), GC (rs2282679), and VDR (rs2228570) affect the concentration of serum 25(OH)D, as an indicator of vitamin D status and play a critical role in the sports performance of CrossFit^® practitioners. Full article

Accurate estimation of rapeseed yield is crucial for harvesting decisions and improving efficiency and output. Machine learning (ML) models driven by remote sensing data are widely used for yield prediction. This study explores the generality of feature-based rapeseed yield prediction models across different varieties and years. Seven vegetation indices (VIs) and twenty-four texture features (TFs) were calculated from UAV-based imagery. Pearson’s correlation coefficient was used to assess variable sensitivity at different growth stages, and the variable importance score (VIP) from the random forest (RF) model was used for feature selection. Three ML regression methods—RF, support vector regression (SVR), and partial least squares regression (PLSR)—were applied using the single-stage VI, selected multi-stage VI, and multivariate VI-TFs for yield prediction. The best yield model was selected through cross-validation and tested for temporal fit using cross-year data. Results showed that the multi-stage VI and RF model achieved the highest accuracy in the training dataset (R² = 0.93, rRMSE = 7.36%), while the multi-stage VI and PLSR performed best in the test dataset (R² = 0.62, rRMSE = 15.20%). However, this study demonstrated that the addition of TFs could not enhance the robustness of rapeseed yield estimation. Additionally, the model updating strategy improved the RF model’s temporal fit, increasing R² by 25% and reducing the rRMSE to below 10%. This study highlights the potential of the multi-stage VI for rapeseed yield prediction and offers a method to improve the generality of yield prediction models over multiple years, providing a practical approach for meter-scale yield mapping and multi-year prediction. Full article

Accurate building energy consumption prediction is essential for efficient energy management and energy optimization. This study utilizes bidirectional long short-term memory (BiLSTM) to automatically extract deep time series features. The nonlinear fitting and high-precision prediction capabilities of Random Forest (RF) and XGBoost models are then utilized to develop a BiLSTM-RF-XGBoost stacked hybrid model. To enhance model generalization and reduce overfitting, a Bayesian algorithm with an early stopping mechanism is utilized to fine-tune hyperparameters, and strict K-fold time series cross-validation (TSCV) is implemented for performance evaluation. The hybrid model achieves a high TSCV average R² value of 0.989 during cross-validation. When evaluated on an independent test set, it yields a mean square error (MSE) of 0.00003, a root mean square error (RMSE) of 0.00548, a mean absolute error (MAE) of 0.00130, and a mean absolute percentage error (MAPE) of 0.26%. These values are significantly lower than those of comparison models, indicating a significant improvement in predictive performance. The study offers insights into the internal decision-making of the model through SHAP (SHapley Additive exPlanations) feature significance analysis, revealing the key roles of temperature and power lag features, and validating that the stacked model effectively utilizes the outputs of base models as meta-features. This study makes contributions by proposing a novel hybrid model trained with Bayesian optimization, analyzing the influence of various feature factors, and providing innovative technological solutions for building energy consumption prediction. It also provides theoretical value and guidance for low-carbon building energy management and application. Full article