Search Results (568)

Explainable artificial intelligence (X-AI) techniques aim to make the actions and decisions of autonomous systems understandable to humans interacting with these systems. In human–AI teams, explainability supports individual understanding and coordination, shared mental models, and collective decision-making among humans and AI agents. Research has shown that X-AI enhances trust in autonomous systems, improves human–AI team performance, and supports collaboration across domains including aviation, finance, healthcare, hospitality, and sports. However, X-AI technologies face difficult challenges, including a lack of transparency and interpretability due to complex underlying models, also known as the “black-box” nature of AI systems. These technologies also lack any universally accepted evaluation metrics and have limited generalizability across applications. One deficit in the X-AI literature is that most frameworks focus on individual-level outcomes, with limited attention to team-level processes. The current study conducted a systematic literature review adhering to PRISMA guidelines and the SALSA framework. This study introduces the Implementation-Design (I-D) framework that organizes X-AI approaches along two dimensions: implementation, ranging from visual to interactive approaches, and design, ranging from isolated explanations to workflow-integrated systems. This framework captures lower-level engagement, involving individual users, to higher-level understanding that is necessary for teams and collectives. Findings indicate that visual explanation approaches support user engagement, while interactive workflow approaches promote deeper understanding, appropriate reliance, and distributed cognition within human–AI teams. Implications highlight the need for team-oriented explainability grounded in shared mental models, transactive memory systems, and collaborative X-AI artifacts. Practical guidelines are included to support researchers and practitioners in selecting appropriate X-AI techniques based on their context and level of analysis. The I-D framework is offered as a conceptual organizing model to guide research and practice, and empirical validation is identified as a priority for future work. Full article

Accurate boxing action recognition is critical for intelligent combat training, action quality assessment, and sports injury prevention. However, existing deep learning approaches face three key challenges: limited feature extraction for high-speed non-rigid boxing motions, weak robustness against background interference and occlusion, and performance instability from labor-intensive manual hyperparameter tuning. Furthermore, the original Centered Collision Optimizer (CCO), a biomimetic algorithm inspired by celestial collision dynamics, suffers from insufficient population diversity, poor adaptive regulation, and premature convergence in high-dimensional hyperparameter optimization tasks. To address these issues, this paper proposes a novel biomimetic optimization-driven boxing action recognition framework, where an Enhanced Adaptive Centered Collision Optimizer (EACCO) automatically optimizes the hyperparameters of a Multi-Scale Spatio-Temporal Adaptive ConvNeXt (MSTA-ConvNeXt) network. First, the MSTA-ConvNeXt backbone integrates multi-scale dynamic deformable convolution, a Bi-GRU spatio-temporal fusion module, and a dual-channel attention mechanism to enhance fine-grained feature extraction and temporal modeling. Second, three biomimetic improvements are introduced to CCO: Tent chaotic elite opposition-based initialization, adaptive nonlinear convergence factor with dynamic weight guidance, and adaptive Gaussian-Cauchy hybrid mutation, which balance exploration and exploitation and avoid local optima. Experiments on two public benchmark datasets show that the proposed framework achieves 96.1% accuracy, 95.9% precision, 95.7% recall, and 95.8% F1-score on the Boxing Jab Skeleton Dataset, and 95.4% accuracy, 95.2% precision, 94.9% recall, and 95.0% F1-score on the Olympic Boxing dataset, outperforming all state-of-the-art methods. Ablation studies validate the effectiveness of each EACCO component and confirm that this biomimetic hyperparameter optimization approach outperforms manual tuning and other popular optimizers. This work provides an effective biomimetic optimization solution for intelligent sports action recognition. Full article

Background: Team sports often involve high-intensity unilateral actions that can lead to neuromuscular asymmetries, increasing injury risk and reducing performance, particularly in young female athletes. Methods: This study quantified and compared inter-limb asymmetries in single-leg countermovement jumps (slCMJ) across sexes and sports in 96 youth elite athletes (16.42 ± 1.03 years; 1.83 ± 0.09 m; 74.36 ± 8.69 kg) from basketball, handball, and volleyball. Using a force plate, asymmetries were assessed, and statistical analyses (t-test, ANOVA) identified differences. Results: Females showed greater asymmetry in jump height (9.64 ± 6.43% vs. 6.48 ± 4.87%, p = 0.01, d = 0.59), whereas males exhibited higher asymmetry in time to take-off (10.32 ± 7.5% vs. 6.4 ± 4.7%, p = 0.003, d = 0.63). Volleyball players displayed the lowest asymmetry in jump height (7.05 ± 4.9%) compared to basketball (8.73 ± 7.2%) and handball (11.88 ± 9.7%, p = 0.05), and in relative maximum power (4.63 ± 3.7%) compared to basketball (7.75 ± 5.3%, p = 0.04) and handball (6.51 ± 4.7%). Conclusions: These findings highlight sex- and sport-specific neuromuscular asymmetry patterns, emphasizing their relevance for injury prevention and performance strategies. However, asymmetries are highly variable, influenced by multiple factors. Full article

Recent advances in wearable, vision-based, trajectory, physiological, and multimodal sensing technologies, together with deep learning, have enabled continuous, objective, and individualized assessment of sport performance and athlete health. Unlike prior reviews that primarily focus on a single sensing modality, sport, or algorithmic series, this review integrates wearable, vision-based, trajectory, physiological, and multimodal sensing streams with deep learning models across both performance analysis and athlete health monitoring, thereby clarifying modality-task-model relationships and translational limitations. This review synthesizes recent progress in sensor-based sports intelligence, focusing on how heterogeneous data streams are transformed into performance- and health-related decision support. The reviewed applications include athlete and ball perception, multi-object tracking, pose estimation, action recognition, trajectory and tactical analysis, training-load and fatigue monitoring, injury-risk prediction, rehabilitation monitoring, and return-to-play support. Deep learning architectures, including CNNs, LSTMs, GRUs, TCNs, Transformers, attention mechanisms, graph neural networks, and multimodal fusion models, are discussed in relation to their suitability for visual, temporal, spatial, physiological, and multisource data. This review further identifies key challenges, including data heterogeneity, annotation scarcity, limited cross-sport and cross-device generalization, real-time deployment constraints, model interpretability, privacy protection, and ethical governance. Moving forward, research efforts should focus on the development of standardized datasets, reliable multimodal data fusion strategies, self-supervised and transfer learning approaches, and deployment on edge or cloud computing platforms. Additionally, enhancing interpretability through explainable AI and implementing closed-loop, individualized monitoring systems are critical. By synthesizing advances in sensing technologies, deep learning methodologies, and real-world applications, this review aims to provide a practical reference for optimizing athletic performance, preventing injuries, guiding rehabilitation, and supporting long-term health management of athletes. Full article

This paper presents an end-to-end system for the automatic summarization of football matches through temporally accurate event spotting across six action categories: goal, card, substitution, kick-off, direct free-kick, and corner. These six classes were selected from the SoccerNet-v2 annotation taxonomy as a non-standard evaluation subset: the three standard SoccerNet-v2 classes (goal, card, substitution) are retained and extended with three additional context-dependent set-piece classes (kick-off, direct free-kick, and corner) that represent the primary unresolved challenge in temporal action spotting. We developed and empirically evaluated a framework based on a Temporal Convolutional Network (TCN) with capsule-based relational encoding applied to class-specific Time-Shift Encoded (TSE) labels using a composite segmentation-detection loss. Evaluated on the SoccerNet-v2 benchmark (300/100/100 train/validation/test splits) using the official PCA-compressed ResNet-152 features sampled at 2 FPS, the implemented TCN-Capsule baseline achieved a test-set mean Average Precision (mAP) of 58.16% at a 10 s tolerance threshold (δ = 20 frames), with a validation mAP of 58.55%, indicating strong generalization with only a 0.39 pp performance gap. The six per-class AP values underlying this result are Goal: 75.6%, Corner: 78.0%, Substitution: 56.9%, Card: 53.9%, Direct Free-kick: 52.8%, and Kick-off: 34.3%; the headline mAP is computed by the official SoccerNet area-under-precision-recall-curve script and is not a simple arithmetic average of these six values. In addition, five complementary evaluation metrics are introduced: Average Temporal Error (ATE), Temporal Recall@K (TR@K), Latency-to-Event (L2E), Temporal Causality Consistency (TCC), and Annotation Noise Sensitivity. The analysis reveals that visually salient events such as goals and corners are detected with high accuracy (AP > 75%, ATE < 5 s, AUC > 0.95 at frame-level), whereas context-dependent events, including kick-offs and direct free-kicks, remain challenging (AP < 55%, ATE > 8 s, AUC = 0.68–0.72). As future work, two architectural extensions are formally specified but not empirically evaluated: a Temporal Segment Network with Non-Local Blocks and a hybrid Transformer-CNN framework with optical flow fusion. Furthermore, an OpenCV-based highlight generation prototype is implemented; formal quality evaluation is planned as future work. Full article

Higher Education Institutions (HEIs) are increasingly tasked with fostering pro-environmental agency to address the global climate crisis. Within physical activity and sport sciences, future nautical sports managers occupy a unique role as stewards of marine ecosystems. This study investigates the sustainability consciousness of 170 undergraduate university students from a single institution. It focuses on analyzing the knowledge–action gap within the framework of the 2030 Agenda. Utilizing the Sustainability Consciousness Questionnaire (SCQ-S), this work assessed knowledge, attitudes, and behaviors to quantify this dissonance through a robust non-parametric statistical pipeline. Results indicate a significant knowledge–action gap ($r_B$ = 0.903), representing a large matched-pairs rank-biserial correlation effect size for the Wilcoxon test. Furthermore, Kruskal–Wallis analyses reveal that this discrepancy is transversal across student profiles, suggesting a state of informed paralysis within this specific cohort, where high theoretical mastery fails to translate into sustainable professional routines. Crucially, the frequency of engagement in nature-based sports, encompassing both nautical and terrestrial outdoor activities, does not significantly reduce this dissonance. These contextual findings indicate that physical exposure to nature does not automatically mitigate the knowledge–action gap, highlighting the need for explicit reflective pedagogies in higher education. Full article

Background/Aim: Motor skill acquisition is the foundation of athletic performance, from the novice learning a new technique to the elite athlete executing complex movements automatically under pressure. Although classical models have defined the neural substrates of motor control—the cerebellum for error correction, the basal ganglia for action selection, and the primary motor cortex (M1) for execution—emerging evidence suggests that motor learning is the result of the dynamic interaction of multiple parallel processes rather than a linear hierarchy. This narrative review integrates classical neuroanatomical knowledge with contemporary findings on multisite plasticity, with a particular focus on sport-specific adaptations. Methods: We examined three core learning mechanisms operating in parallel: error-based learning (cerebellar-dependent, driven by sensory prediction errors), reinforcement learning (striatal-dependent, driven by reward prediction errors and dopamine), and use-dependent learning (cortical-dependent, driven by mere repetition). We also summarize the biochemical pathways supporting these learning processes, including glutamatergic LTP-like cortical plasticity, cerebellar mGluR1–PKC–LTD signaling, dopaminergic corticostriatal plasticity, BDNF–TrkB-dependent neurotrophic mechanisms, growth-factor signaling, and exercise-induced muscle–brain communication. Results: We then propose a spatiotemporal model in which the relative contribution of each network shifts dynamically across the three stages of skill acquisition, from the early cognitive/strategic phase to the late automatic phase characteristic of elite performance. At the molecular level, these stage-dependent adaptations are supported by synaptic strengthening and weakening mechanisms, reward-dependent dopamine signaling, neurotrophic and growth-factor-mediated remodeling, and peripheral metabolic/myokine signals that modulate brain plasticity during training and recovery. Special attention is given to contextual and sport-specific adaptations, using the paradigmatic example of elite swimmers who demonstrate enhanced short-interval intracortical inhibition (SICI) selectively in the aquatic environment, reflecting long-term sport-induced neuroplasticity. Conclusions: Understanding these dynamic network mechanisms has direct implications for coaching, training periodization, and the development of targeted neuromodulatory interventions to accelerate skill acquisition and optimize athletic performance. Full article

Action quality assessment (AQA) is vital in sports science and clinical rehabilitation, but existing methods suffer from data noise sensitivity, weak physical interpretability, and limited generalization across body types. This study proposes a novel AQA paradigm for standing long jump to address these limitations. A methodological framework based on kinematic reconstruction and physical deviation analysis is developed. Ideal trajectories derived from forward kinematics establish a unified evaluation benchmark. Three deviation features (Peak Deviation, Cumulative Deviation Integral, Phase-Weighted Deviation) are computed between ideal and experimental samples, then input into a Random Forest Classifier for automatic grading. Experimental results demonstrate the proposed method significantly outperforms traditional dynamic-time-warping-based templates in accuracy and robustness. Ablation studies confirm kinematic trajectory benchmarks are superior to raw or averaged trajectories. Model decisions show strong consistency with physical characteristics, verifying interpretable assessment capability. This work shifts AQA from “data similarity comparison” to “physical rationality evaluation”, offering a new perspective for vision-based automated motion analysis. Full article

Bilateral deficit (BLD) is traditionally defined as the reduced capacity to produce force during simultaneous bilateral contractions compared with the summed output of unilateral actions. However, in applied sport settings, BLD is frequently estimated from countermovement jump (CMJ) performance, representing a functional rather than a direct mechanical measure of force production. Therefore, the aim of this study was to examine the association between a CMJ-derived functional BLD index and multidirectional performance in soccer players. Forty male university soccer players (age: 23 ± 1 years) performed unilateral and bilateral CMJ. The BLD index was calculated from jump height values obtained during these assessments. Participants subsequently completed the 505 change-of-direction (CoD) test, which was analyzed using two-dimensional video-based motion analysis. Participants were classified according to BLD magnitude into low, moderate, and high BLD groups. Group differences were assessed using Kruskal–Wallis tests with Bonferroni-adjusted post hoc comparisons. Additionally, Spearman correlation analyses were performed using BLD as a continuous variable. Significant between-group differences were observed across all temporal phases of the 505 test (p < 0.001), with players exhibiting lower BLD values demonstrating superior acceleration, deceleration, reacceleration, and overall CoD performance. Significant negative correlations were also observed between BLD and reaction time, acceleration, deceleration, reacceleration, CoD time, and CoD deficit (r_s = −0.42 to −0.69; p < 0.001). No significant associations were found for stride length, acceleration ability, or inter-limb asymmetry. These findings suggest that lower magnitudes of a CMJ-derived functional BLD index are associated with superior multidirectional performance in soccer players. However, given that BLD was estimated from jump performance, the results should be interpreted as associations with a functional neuromuscular performance index rather than as direct evidence of bilateral force production capacity. Full article

Although athlete monitoring can quantify training exposure and athlete status with increasing detail, conversion into daily training decisions remains inconsistent. This structured narrative review synthesizes evidence on training load, neuromuscular readiness, recovery, fatigue interpretation, measurement reliability, applied decision-making, and proposes the LOAD-R framework: a systems model linking Load, Organism response, Adaptive state, Decision, and Re-evaluation. A transparent non-PRISMA strategy was used because the aim was conceptual integration and framework development rather than effect-size pooling. Evidence was organized around field-applicable monitoring domains and their decision value. LOAD-R builds on existing monitoring approaches by organizing single indicators, fixed thresholds, and dashboard alerts into an explicit interpretation-to-action sequence. It classifies athlete state into adaptive, functional-overload, underloaded, uncertain, or maladaptive zones, each linked to progress, maintain, modify, deload, or recover decisions. The framework also provides implementation levels and testable predictions. By framing monitoring as adaptive decision support rather than passive data collection, LOAD-R may improve decision consistency, reduce maladaptive training responses, and enhance the practical value of athlete monitoring in applied sport settings. Full article

Background: Football is an intermittent sport characterized by high physical and physiological demands, which may be influenced by the competitive level. Understanding differences in match load is fundamental for optimizing training planning, fatigue management, and athlete performance and injury prevention. This study aimed to evaluate and compare external and internal load in senior football players in Portugal across five distinct competitive levels. Methods: Wimu Pro^TM (Hudl, Lincoln, NE, USA) and Garmin Heart Rate bands (Garmin International Inc., Olathe, KS, USA) were used to quantify and evaluate the external and internal load of the players. A total of 96 athletes were assessed, with ages ranging from 19 to 36 years (mean: 24.28 ± 4.72), who were divided into five competition levels (1st Division (n = 19), 2nd Division (n = 21), 3rd Division (n = 14), 4th Division (n = 20), and Regional Division (n = 22). Results: Significant differences were observed between competitive levels across several external load variables (p > 0.001). The 3rd Division and 4th Division showed higher values in variables associated with reactive and high-intensity actions (p < 0.001; effect size: 0.287), whereas the 2nd Division exhibited a more controlled load profile. Regarding internal load, significant differences were only observed in average heart rate during the second half (p = 0.043; effect size: 0.085), indicating distinct capacities to maintain physiological intensity under fatigue. Conclusions: It can be concluded that competitive level influences load profiles in football, although the differences do not follow a linear pattern. External and internal loads demonstrate greater discriminatory capacity between competitive levels than internal load. Full article

Background/Objectives: Curcumin has been proposed as a nutritional strategy to support exercise recovery through antioxidant and anti-inflammatory actions. However, trials differ in sport context, training status, supplementation timing, dose, formulation, and methodological control. This systematic review evaluated its effects on recovery outcomes in active individuals and athletes, with particular attention to the applicability of the evidence to real-world sport settings. Methods: PubMed, Scopus, Web of Science, SPORTDiscus, and Cochrane Library/CENTRAL were searched from 2012 to June 2026. Randomized double-blind placebo-controlled trials were eligible when they evaluated oral curcumin, curcuminoids, Curcuma-derived preparations with a specified curcumin dose, or curcumin combined only with bioavailability enhancers. Studies using artificial muscle-damage protocols, clinical populations, non-randomized designs, or combined bioactive interventions were excluded. Methodological quality was assessed using the Physiotherapy Evidence Database (PEDro) scale, supplemented by a Cochrane Risk of Bias 2 (RoB 2) assessment and a Grading of Recommendations Assessment, Development and Evaluation (GRADE) certainty-of-evidence evaluation. Owing to heterogeneity, findings were synthesized narratively by outcome domain, supplementation timing, formulation type, exercise context, and training status. Results: Fifteen trials were included. Favorable effects were reported in 6/7 studies assessing oxidative stress, 4/6 assessing muscle damage, 3/8 assessing inflammation, 3/7 assessing subjective recovery, soreness, or fatigue, and 4/8 assessing physical or athletic performance. However, effects varied substantially according to population, exercise context, biomarker selection, timing of assessment, and formulation type. The certainty of evidence was low for oxidative stress and very low for muscle damage, inflammation, subjective recovery/soreness/fatigue, and performance. Conclusions: Curcumin supplementation may support selected aspects of exercise recovery, particularly oxidative stress responses. However, these findings should be interpreted cautiously because the evidence derives mostly from small trials with heterogeneous populations, exercise protocols, supplementation regimens, formulations, biomarkers, and assessment time points. Evidence for muscle damage, inflammation, subjective recovery, fatigue, and performance remains inconsistent, and further well-controlled trials in trained and high-performance athletes are needed before practical recommendations can be established. Full article

European catfish, an invasive alien species (IAS) that inhabits the Tagus River in Extremadura, is a threat to freshwater biodiversity and the local economy. As a freshwater mega-predator, it eats all kinds of fish, impacting local ichthyofauna as well as sport-recreational angling, a major economic driver in Extremadura, with 100,000 angling licenses issued by the Extremadura local government and more than 1000 annual bank fishing competitions. The introduction of Silurus glanis into Tagus basin is recent, dating to the 20th century, but it has rapidly spread. Although this expansion is driven by anglers, boat anglers account for only two per thousand anglers in the region, and even among them European catfish anglers are scarce, as fishing for this species is prohibited in the region and local anglers and anglers associations are against European catfish expansion. Efforts to control catfish in Extremadura have been based on studies of selective capture methods, such as trammel nets, gill nets, long lines, fish traps, and electrofishing combined with nets in areas where barbel congregate during their spawning migration. The most effective and selective methods for European catfish, avoiding bycatch, are trammel nets and electrofishing. Therefore, a combination of these methods and collaboration between local government and anglers are proposed for its control in reservoirs. In cases of recent introductions into standing water, rapid actions have been taken, involving complete drainage of the water body, rescue of native fish and removal of European catfish. These measures are feasible in reservoirs up to 10 hectares and can be combined with sediment management to improve water quality. Deterrence and control measures have been implemented in response to the promotion of catfish angling in regional media and on social media. To control European catfish expansion, collaborative actions involving public agencies, local government, angler associations and public awareness are essential. Full article

The integration of wearable inertial sensors into sports performance analysis has enabled the automatic quantification of complex movement patterns, particularly in highly specialized roles such as the goalkeeper. However, the ability of sensor-derived metrics to characterize and differentiate goalkeeper-specific actions across different training contexts remains limited. This pilot study adopted a data-driven, multilevel approach to analyze external load and specific actions in a semi-professional female football goalkeeper across 39 training sessions comprising 148 tasks. The aims were: (i) to characterize external load profiles according to task type, (ii) to identify differences in mechanical demands across training contexts, and (iii) to determine the latent structure of goalkeeper-specific load. Descriptive analyses, coefficient of variation (CV), Kruskal–Wallis tests with post hoc comparisons, and principal component analysis (PCA) were conducted. The results revealed a clear task-dependent load gradient, with goalkeeper-specific tasks showing higher values than small-sided games and full-game situations, particularly for variables such as dives/min, dives load/min, and impacts +8/min. High variability was observed, with coefficient of variation values exceeding 100% in several indicators, reflecting the intermittent nature of goalkeeper involvement. Principal component analysis identified a two-component structure explaining 76.6% of the total variance, associated with specific involvement intensity and postural-control/movement-load demands. In this single-case pilot study, the analyzed goalkeeper’s external load was task-dependent, highly variable, and multidimensional. In comparable women’s football contexts, these findings support the use of task-specific inertial monitoring to verify and adjust female goalkeepers’ exposure to key position-specific actions, such as dives, impacts, accelerations/decelerations, and set-position demands, rather than relying exclusively on full-game tasks or generic locomotor indicators. Full article

Background/Objectives: In interceptive sports like volleyball, the ability to accurately time an action relative to a moving object (Coincidence Anticipation Timing—CAT) is critical. This study investigated the effects of a structured 8-week deliberate practice intervention on CAT accuracy and its long-term retention in young female athletes. Methods: Thirty-two female volleyball players (aged 8–10 years) were randomly assigned to an Experimental Group (EG, n = 16) and a Control Group (CG, n = 16). The EG underwent a specialized 8-week training program focusing on progressive cognitive load and immediate knowledge of results, while the CG followed standard volleyball training. A single-blind assessor measured CAT at two velocities (5 mph and 10 mph) using the Bassin Anticipation Timer at three time points: pre-test, post-test, and a 2-month retention test. Results: ANCOVA revealed a significant group-by-time interaction (p < 0.001), with the EG demonstrating a substantial reduction in absolute timing error. The effect size was markedly higher at 10 mph (partial η² = 0.400) compared to 5 mph (partial η² = 0.197). Crucially, the EG maintained their performance gains during the retention test (p < 0.05), whereas the CG showed no significant improvement over time. Conclusions: Targeted deliberate practice effectively enhances temporal prediction accuracy in children, likely by facilitating a shift from reactive to predictive motor control. The robust retention of these skills underscores the significant neuroplasticity of the 8–10 age window, suggesting that early foundational timing interventions provide long-lasting benefits for athletic development. Full article