icSPORTS 2025 Abstracts

Area 1 - Computer Systems in Sports

Full Papers

Paper Nr:	15
Title:	Cognitive Load and Motor Adjustment Under Virtual Defensive Pressure in Mixed Reality Sports Training
Authors:	Hayato Saiki, Kiyohiro Konno, Kosuke Naruse, Takahiro Shimizu, Yasuhiro Suzuki, Seiji Ono and Kenji Suzuki
Abstract:	Defensive pressure experienced during competition exerts a significant impact on athletes’ psychological states and motor behaviors; however, reproducing such pressure in training environments remains a persistent challenge. This study developed a Mixed Reality (MR)-based training system that enables athletes to experience realistic defensive pressure during actual physical movements, using virtual avatars that perform closeout actions. By integrating motion capture technology with an optical see-through head-mounted display, we constructed an environment where the perceived intensity of defensive pressure could be modulated by manipulating avatar height. Experiments involving nine skilled male basketball players compared three defensive conditions-Free (no defender), Human defender, and Virtual defender (180 cm)-and further examined the effects of avatar height variations (160 cm vs. 200 cm). Results indicated that virtual defenders, despite the absence of physical contact, induced psychological load comparable to that of human defenders, as measured by NASA-TLX scores and ball-holding duration. Notably, taller avatars elicited greater perceived pressure and accelerated shot preparation, although no significant differences in shooting accuracy were observed. These findings suggest that MR-based training systems offer an effective means to systematically scale perceived difficulty without compromising performance. Furthermore, the observed dissociation between subjective workload and final outcomes highlights the importance of incorporating multisensory feedback to enhance ecological validity in future system designs.

Paper Nr:	17
Title:	Edging Velocity: The Crucial Role of Edge Engagement in Alpine Skiing
Authors:	Christoph Thorwartl, Thomas Grah, Harald Rieser, Günter Amesberger, Stefan Kranzinger, Thomas Stöggl, Helmut Holzer and Thomas Finkenzeller
Abstract:	In alpine skiing, the way a ski engages with the snow surface – particularly at the beginning of a turn – plays a key role in determining performance. This study introduces Edging Velocity (EV) as a novel metric to quantify how quickly the ski is tipped onto its edge during turn initiation. Building upon sensor-based motion analysis using the “Connected Boot” system, we investigated three distinct skiing techniques: race carving, moderate carving, and parallel ski steering. An expert skier performed multiple turns for each technique, and EV was computed from edge angle progression. Results show that EV was highest during race carving, followed by moderate carving, and lowest during parallel ski steering. All pairwise differences were statistically significant (p < 0.001 or p < 0.01). These findings highlight EV’s potential as a performance-relevant parameter for optimizing edge engagement. Integrated into real-time feedback systems, EV may support learning and refinement of skiing technique, particularly in the critical early phase of a turn.

Paper Nr:	21
Title:	Can We Really Predict Which Football Players Will Succeed?
Authors:	Jonathan Feldman
Abstract:	Predicting whether a football player will achieve their projected career potential is a key challenge for clubs and scouts. This study analyzes the career outcomes of 8,770 players from the European Soccer Database (2008–2016), using FIFA video game potential ratings as a proxy for projected potential. To account for the increasing difficulty of skill improvements at higher levels, we apply logarithmic scaling when calculating achievement ratios. Predictive models were trained on two cohorts: players with complete career data and those with early-career data (up to age 21). Early-career models achieved moderate predictive performance (ROC AUC = 0.79), reflecting the challenge of identifying long-term success based on limited early observations. SHAP analysis shows that growth trajectory features, including early improvement and development patterns, contribute more to success predictions than static physical or technical attributes. We define success as fulfilling projected potential according to the FIFA rating system – a standardized but subjective benchmark. While this does not capture all real-world outcomes, it enables large-scale analysis of developmental trajectories. These results suggest that tracking player development over time provides better guidance for talent decisions than relying solely on early physical assessments.

Paper Nr:	23
Title:	Balance of Upper Limb Muscle Activation and Aerodynamics for Cycling Posture Optimization
Authors:	Xiangru Li, Peng Zhou and Xin Zhang
Abstract:	Cycling postures significantly influence aerodynamic performance in competitive cycling, yet aggressive postures may increase muscle activation and lead to adverse physiological effects. This study evaluated the aerodynamic drag area (CdA) of various cycling postures in a wind tunnel, revealing a strong correlation with decreasing forearm angles. Surface electromyography (sEMG) tests were conducted to assess upper limb muscle activation across postures, identifying the triceps brachii (TB) as dominant in maintaining both hoods and drops positions (46.9% and 40.2% of total activation, respectively). Additionally, this study explores the trade-off between aerodynamic gains and muscle activation by examining the relationship between CdA and composite EMG. A Pareto front analysis identified locally optimal postures that balance these factors, potentially enhancing overall cyclist performance.

Paper Nr:	25
Title:	S-amba: A Multi-View Foul Recognition in Soccer Through a Mamba-Based Approach
Authors:	Henry O. Velesaca, Alice Gomez-Cantos, Abel Reyes-Angulo and Steven Araujo
Abstract:	In this work, we propose a novel Mamba-based multi-task framework for multi-view foul recognition. Our approach leverages the Mamba architecture’s efficient long-range dependency modeling to process synchronized multi-view video inputs, enabling robust foul detection and classification in soccer matches. By integrating spatial-temporal feature extraction with a multi-task learning strategy, our model simultaneously predicts foul occurrences, identifies foul types, and localizes key events across multiple camera angles. We employ a hybrid loss function to balance classification and localization objectives, enhancing performance on diverse foul scenarios. Extensive experiments on the SoccerNet-MVFoul dataset demonstrate our method’s superior accuracy and efficiency compared to traditional CNN and Transformer-based models. Our framework achieves competitive results, offering a scalable and real-time solution for automated foul recognition, advancing the application of computer vision in sports analytics. The codebase is publicly available at https://github.com/areyesan/Mamba-Based MVFR for reproducibility.

Paper Nr:	35
Title:	Assessing and Visualizing Principles of Play in Soccer
Authors:	Rúben Filipe Rocha and Rui J. Lopes
Abstract:	Recent advances in soccer analytics have significantly expanded the use of spatial and temporal data to un-derstand tactical behaviour. However, many visualisation tools remain limited in their ability to contextualize these behaviours according to established principles of play. This paper presents a context-aware visualisation framework that leverages StatsBomb Open Data to identify, illustrate, and interpret tactical patterns in professional soccer. The tool enables analysts to explore match phases through key tactical dimensions such as compactness, pressing, width, support, and penetration, using event-level data enriched with positional data (360 freeze frames). Unlike generic dashboards or statistical summaries, the proposed system integrates spatial relationships and collective movements, offering a more accurate representation of team behaviour across different match contexts. By combining visual methods with tactical theory, the tool supports coaches and analysts in identifying the match principles of play, thereby facilitating a deeper understanding of performance dynamics.

Paper Nr:	54
Title:	Revisiting Expected Possession Value in Football: Introducing a U-Net Architecture, Reward and Risk for Passes, and a Benchmark
Authors:	Thijs Overmeer, Tim Janssen and Wim P. M. Nuijten
Abstract:	This paper presents an Expected Possession Value (EPV) model for football with three main new components: a U-Net-inspired convolutional neural network architecture, ball height as a feature, and a dual-component pass value model that analyzes reward and risk. We furthermore introduce the Overmeer–Janssen–Nuijten Pass Expected Possession Value benchmark (OJN-Pass-EPV benchmark), which enables a quantitative evaluation of EPV models by using pairs of game states with given relative EPVs. The presented EPV model achieves good results in model loss and Expected Calibration Error on a dataset containing Dutch Eredivisie and 2022 FIFA Men’s World Cup matches and correctly identifies the higher value state in 78% of the game state pairs in the OJN-Pass-EPV benchmark, demonstrating its ability to accurately assess goal-scoring potential. Our findings enable more precise EPV estimations, support risk-reward analysis for passing decisions, and establish quality control standards for EPV models.

Short Papers

Paper Nr:	14
Title:	From Marketing to the Court: Applying MMM and Fatigue Analysis for Optimal Basketball Lineups
Authors:	Nachi Lieder
Abstract:	Lineup optimization in basketball is crucial for maximizing team performance. Traditional methods often overlook the effects of player fatigue and the absence of historical data for certain lineups. This study introduces an advanced model inspired by marketing mix models (MMM), which incorporates player fatigue to optimize lineups and maximize the plus-minus metric. By examining two distinct lineups: In one high-tempo, fast-paced, and the other slow-paced, and durable, we highlight the varying impacts on productivity and fatigue. The fast-breaking lineup may show higher immediate productivity but suffers from quicker fatigue, while the durable lineup maintains consistent performance over a longer period.

Paper Nr:	20
Title:	Gait-Based Prediction of Penalty Kick Direction in Soccer
Authors:	David Freire-Obregón, Oliverio J. Santana, Javier Lorenzo-Navarro, Daniel Hernández-Sosa and Modesto Castrillón-Santana
Abstract:	Understanding and predicting penalty kick outcomes is critical in performance analysis and strategic decision-making in soccer. This study investigates the potential of gait-based biometrics to classify the intended shoot zone of penalty takers using temporal gait embeddings extracted from multiple state-of-the-art gait recognition backbones. We compile a comprehensive evaluation across several models and datasets, including baseline models and other models such as GaitPart, GLN, GaitSet, and GaitGL trained on OUMVLP, CASIA-B, and GREW. A standardized LSTM-based classifier is trained to predict the shooting zone from video-level gait sequences, using consistent train-test splits to ensure fair comparisons. While performance varies across model-dataset pairs, we observe that certain combinations yield better predictive accuracy, suggesting that the gait representation and the training data influence downstream task performance to some degree. This work demonstrates the feasibility of using gait as a predictive cue in sports analytics. It offers a structured benchmark for evaluating gait embeddings in the context of penalty shoot zone prediction.

Paper Nr:	24
Title:	Physical and Physiological Impact of Custom-Made Virtual Reality Exergames for Older Adults
Authors:	Cíntia França, Hildegardo Noronha, Eva Freitas, Pedro Campos, Rui T. Ornelas and Élvio R. Gouveia
Abstract:	Aging is associated with decreased physical fitness components, such as strength, power, cardiorespiratory fitness, and balance, resulting in physical limitations on functional activities of daily living. The increasing population of older adults calls for innovative strategies to support functional health, particularly through exercise. This study has two main objectives: (1) to introduce a custom-designed virtual reality (VR) exergame prototype, FitFest, developed to deliver physical activity (PA) sessions for older adults, and (2) to present the results of a pilot study assessing physical and physiological responses during gameplay. Seven older adults (mean age 67.0±3.8 years) participated in 18 user testing sessions involving two VR exergames—Wine Fest and Flower Fest. Each session was monitored for PA intensity and heart rate (HR). The participants spent most of their time in sedentary behavior (56.5±20.4%), followed by light PA (42.1±19.3%), averaging 436.7 steps and a heart rate of 92.1 bpm per session. Although the differences were not statistically significant, Wine Fest led to lower sedentary behavior, higher light PA levels, and more total steps than Flower Fest. The findings suggest that the system can potentially promote light PA among older adults, emerging as a complementary tool to traditional PA sessions.

Paper Nr:	30
Title:	Deep Learning-Based Autoencoder for Objective Assessment of Taekwondo Poomsae Movements
Authors:	Mohamed Amine Chaâbane, Imen Ben Said and Adel Chaari
Abstract:	Artificial Intelligence (AI) is revolutionizing sports by enhancing performance, improving safety, and creating richer fan experiences. This paper focuses on leveraging AI in Taekwondo, specifically in assessing athlete performance during Poomsae movements, which are foundational to the sport and crucial for success in competitions. Traditionally, the evaluation of Poomsae has been subjective and heavily reliant on human judgment. This study addresses this issue by automating the assessment process. We propose a deep learning approach that utilizes computer vision to analyse athletes' movements captured in video clips of Poomsae. The proposed approach is based on a model that emphasizes the use of autoencoders for training data representing skeleton body points of correct movements. This model can effectively identify anomalies, i.e., incorrect movements by athletes. The SportLand platform implements the proposed approach, providing coaches and athletes with precise and actionable insights into their performance. This platform can serve as an assistant for self-evaluation, allowing Taekwondo athletes to enhance their skills at their own pace.

Paper Nr:	36
Title:	Ski Tip Display: Design and Implementation of an Unobtrusive Ski Mounted Visual Feedback System for Alpine Skiing
Authors:	Thomas Grah
Abstract:	This paper presents the design, implementation, and field evaluation of SkiTip Display, a ski-mounted LED feedback system that provides terminal visual feedback on carving angle to support skill development in alpine skiing. Developed through a participatory design process with domain experts, the system was deployed in-the-wild and tested with nine recreational skiers. Results from sensor-based metrics and post-session interviews suggest that ski-mounted visual feedback is perceivable, motivating, and well-suited for post-run reflection, though not actionable during motion. We report key lessons on feedback timing, simplicity, and trust, and discuss implications for designing embedded performance feedback in high-speed outdoor sports. This work contributes to the field by expanding the design space for equipment-integrated feedback systems and by articulating challenges of in-the-wild deployment in dynamic environments.

Paper Nr:	52
Title:	Machine Learning-Based Stroke Segmentation in Kayaking Using Integrated IMU and EMG Data
Authors:	Gábor Nagy, Péter Katona, Levente Gannorouwa and László Grand
Abstract:	Accurate classification of stroke side in rowing motions is essential for performance monitoring and injury prevention. This study evaluates three machine learning models — Naive Bayes (NB), Logistic Regression (LR), and Gradient Boosting Decision Trees (GBDT) — using biomechanical and electromyographic (EMG) features. A core set of 25 features was identified, with normalized joint coordinates and latissimus dorsi EMG activity among the most influential. The NB model achieved 92.21\% cross-validation accuracy using only three coordinate-based features, while the full feature set improved accuracy modestly by 1.94\%. The LR model attained 94.48\% accuracy, slightly outperforming NB. The GBDT model achieved the highest accuracy with 96.18\% on the test set, alongside the lowest mean absolute stroke onset detection error of 24.6 \pm 51.6 ms, corresponding to just 4.5\% of average stroke duration. Classification accuracy remained stable across stroke paces. A strong negative correlation (R = -0.935) between classification accuracy and onset detection error was observed across subjects, indicating that poorer spatial classification corresponds with greater temporal imprecision. Significant inter-subject variability was found, with accuracy ranging from 91.89\% to 98.9\%, likely reflecting individual differences in stroke technique and muscle activation patterns. A core set of biomechanical features were identified, such as normalized joint coordinates of th eulnar styloid and right olecranon, latissimus dorsi EMG activity among the most influential, vertical pelvis lateral bending and bilateral shoulder flexion. Tempo-based relative time averages of these features reveal clear phase-dependent patterns that contribute strongly to model decision-making. These results demonstrate that accurate stroke side classification can be achieved using a relatively small set of biomechanical features, with GBDT models providing superior performance.

Paper Nr:	60
Title:	Assessment of Visuomotor Coordination in Team Sports: Concept and Implementation
Authors:	Anton Ezhov, Anna Zakharova and Kamiliia Vinokurova
Abstract:	Coordination skills are among the main components of success in team sports. Despite the wide development and distribution of methods for coordination development in sport, there is a lack of objective monitoring methods for coordination assessment especially in the context of changes in the terms of game situation. The article presents a unique innovative method for objective coordination assessment based on the team sport training system “Co-Star”. Co-Star includes fixed position of sensors and proprietary activation algorithms. Thus an objective and reliable evaluation of the bond «stimulus - reaction - precise action» is carried out. This evaluation method allows to estimate both athlete’s own (general) visuomotor coordination as well as specific coordination with the use of game equipment (ball or racket). Co-Star system may be useful in training and testing athletes of all ages from young ones to professionals in team sports. The proposed athletes age norms formed as a result of our research potentially would help in sports selection.

Paper Nr:	13
Title:	Step Length Measurement Through Signal Power Analysis and Accelerometer Device: A Laboratory Comparison
Authors:	Christopher S. Ramon, Santiago R. Coronel, Jostin L. Ruiz-Zambrano, Raúl I. Villalta-Encalada, Julio C. Chuqui-Calle, Luis J. Serpa-Andrade, Freddy L. Bueno-Palomeque and Paúl A. Chasi-Pesántez
Abstract:	The monitoring and analysis of kinetic and kinematic parameters integrated into sports disciplines such as race walking can provide valuable information for developing personalized training programs and evaluating technique execution. In this study, we developed two systems for measuring step length based on microcontrollers ESP32. The first system is based on measuring the received signal strength between two antennas, while the second relies on inertial sensors (MPU-9250). Both systems were tested in a laboratory setting using a treadmill and video recording to assess their accuracy. The results showed that the system based on signal strength measurement exhibited low precision at distances within a range of a few centimetres. On the other hand, the inertial sensor-based system demonstrated higher accuracy when compared to video recordings. Although the measurements differed statistically between these two methods (p-value = 0.001), the proposed inertial system recorded a step length of 65 (61-69) cm, while the video recordings measured 67 (64-70) cm. The error distribution analysis showed that 39\% of measurements had an error of 3.2 cm, 32\% had an error of 7.5 cm, and 29\% had an error not exceeding 12 cm. The proposed system shows potential for step length quantification using the MPU-9250 sensor; however, further testing is required to reduce the measurement errors.

Paper Nr:	18
Title:	A Novel Approach to Automated Live-Ticker Generation in Football: Using Large Language Models and Audio Data
Authors:	James Anurathan, Manfred Rössle and Marco Klaiber
Abstract:	Football (soccer) is one of the most popular sports in the world, with fans enjoying real-time coverage of their favorite team’s from anywhere. Explicitly, the progress in the field of Artificial Intelligence (AI) holds great potential to further improve this experience and optimize the delivery of content. In this context, our work investigates the integration of Large Language Models (LLMs) – in our case GPT-4 – with Advanced Speech Recognition (ASR) systems to automate the creation of live football ticker commentary. For this purpose, we present an approach for transcribing live audio commentary from real football matches, utilizing a whisper model to prepare the transcribed text for correct input to the LLM. This approach is leveraged by Named Entity Recognition (NER) and BERT-based models to provide clear, user-friendly, and multilingual texts for live tickers. In addition, we evaluate our approach with an objective and metric-based method to transparently assess the effectiveness of our approach. The study shows the potential of LLMs in automating sports commentary, but also emphasizes the importance of refining entity recognition and addressing content accuracy issues. Future work should focus on improving transcription accuracy, refining NER models, and mitigating LLM hallucinations to develop more reliable and scalable automated live ticker systems.

Paper Nr:	33
Title:	Modern Light Sport Training Systems: “Co-Reaction” Neuromuscular Warm-Up
Authors:	Anton Ezhov and Anna Zakharova
Abstract:	Warm up is an irreplaceable part of sport training. Traditional pre-competition warm-up is fulfilled automatically and therefore do not activate the brain in the way that is needed for excellent performance especially in sprint or first minutes of team sports game. There is no objective assessment of the warming up effectiveness and current state of the athlete. The aim of the study was twofold: to substantiate pre-competition brain warm-up using modern light sport training systems and suggest the method of objective assessment of warm-up efficiency in sport. The innovative Co-Reaction sports system, including the BlazepodTM Trainer with 6 sensors, the BlazepodTM software installed on a smartphone or tablet with Co-Reaction add-ons, and the Co-Reaction Desk, can provide rapid neuromuscular activation to athletes before competition. The results of Co-Reaction (visual-motor two-colour reaction time) allow to assess a current state of an athlete during the pre-competition warm-up. Comparison of the current result with the average earlies ones will give feedback on the athlete readiness/unreadiness and sufficiency/insufficiency of the warm-up. The article presents average data of Co-Reaction time and athletes' norms for different sport and Co-reaction results ages features of ice-hockey players.

Paper Nr:	62
Title:	Esports Impact on School Culture and Climate in K-12 Schools: A Literature Review
Authors:	Alvaro Brito
Abstract:	This literature review explores the impact of esports on school culture and climate, focusing on how it influences student well-being, self-regulation, and psychological development of K-12 students. The literature review examines existing research to understand the implications of esports participation within educational contexts, specifically in elementary and high school grade levels. The review will address key dimensions such as safety, relationships, teaching and learning, institutional environment, and the school improvement process. Additionally, the literature review provides some background on the potential benefits and challenges associated with esports in K-12 schools, shedding light on its effects on students’ health, mindset, and overall school experience. These major groupings were derived from the general inquiry questions on esports school culture and climate from specific research that is currently available.

Area 2 - Health and Support Technology

Full Papers

Paper Nr:	28
Title:	Localized Thermal Analysis for Sportswear via Wind Tunnel Testing
Authors:	Xiaoyi Cai, Zixiang Hu and Peng Zhou
Abstract:	Thermal comfort of sportswear is critical for optimizing athletic performance and improving safety. However, current research on the convective heat transfer coefficient (h) between fabrics and the air flow remains limited, especially in terms of its local analysis. In this study, a thermal cylinder which simulates human limb was developed to evaluate the fabrics' thermal insulation (It) and h (both local and global values) in the wind tunnel. The wind speed ranged from 2 to 8 m/s, and the stretch ratio (SR) from 1.1 to 2.0. The results show that for both fabrics, the local h reaches a minimum around θ = 80°–90°, while the local It peaks near θ = 120°. The effects of SR and wind speed on It and h are also reported. This work offers a practical method for quantifying the heat transfer characteristics of stretched fabrics, providing theoretical guidance for sportswear design and thermal regulation strategies in wind environment.

Short Papers

Paper Nr:	32
Title:	Impact of Physiological Characteristics on Thermal Comfort of Cycling Helmet
Authors:	Zixiang Hu, Xiaoyi Cai and Peng Zhou
Abstract:	Research on helmet thermal comfort is crucial for optimizing helmet design and enhancing cyclists’ acceptance of helmets. However, existing studies often neglect the impact of cyclists’ physiological characteristics on scalp heat dissipation details. To address this gap, a sweating thermal mannequin head was developed to investigate the effects of physiological factors, including hair, sweating, and variations in head pitch angle, on scalp heat dissipation at typical cycling speeds. The findings reveal that hair obstructs airflow within the helmet, resulting in local thermal discomfort at the back of the scalp and potentially altering the optimal pitch angle for helmet thermal performance. Moreover, sweating amplifies the temperature differences between local ”hot spots” and ”cold spots” on the scalp, with the majority of heat loss attributed to sweat evaporation. Additionally, adjusting the pitch angle can better align the ventilation holes with the airflow, thereby enhancing thermal comfort at both the front and back of the scalp.

Paper Nr:	46
Title:	E-Cargo Bikes: Investigation of Innovative Bike Frame Materials - Wood vs. Carbon Fibre and the Impact of Pro-Environmental Attitudes, Age, and Gender
Authors:	Klemens Weigl, Verona Pircher, Justin Hanslmeier and Mira Scheuvens
Abstract:	E-Cargo bikes are often promoted in the context of climate change and sustainability as a great alternative to cars. Even though the number of E-Cargo bikes increases worldwide, it is unclear whether innovative bike frame materials such as wood and carbon fibre may play an important role for interested cyclists. Hence, we conducted an exploratory cross-sectional online questionnaire study and collected data from 292 participants in Germany (147 female, 144 male, one diverse; age range: 18 to 88 years). They completed scales, such as a semantic differential and buying intention, both for E-Cargo bikes made of wood and carbon fibre, respectively, as well as a questionnaire on environmental attitudes. We found that respondents who reported greater values on pro-environmental attitudes favoured wooden E-Cargo bikes. However, we observed no preference between E-Cargo bikes made of wood vs. carbon fibre. Additionally, we uncovered a gender effect for pro-environmental attitudes. Therefore, we conclude that cyclists do not prefer wood or carbon fibre as a bike frame material for E-Cargo bikes. Women seem to be more interested in wooden E-Cargo bikes than men, while articulating more pro-environmental attitudes that could impact environmental campaigns on green mobility.

Paper Nr:	42
Title:	Digital Phenotyping and Behaviour Change in Addiction Recovery: Towards Personalized Physical Activity Interventions
Authors:	Pedro Morouço, Tânia Caetano and Eduardo Ramadas
Abstract:	This position paper explores the integration of digital phenotyping into addiction recovery through a 3-in-1 mobile application that combines adaptive physical activity, psychological support, and mindfulness. Framed within the clinically applied Change & Grow® Therapeutic Model, the proposal aims to transform real-time behavioural and physiological data into personalized, responsive interventions for individuals recovering from substance use disorders. The primary research objective is to assess whether an intelligent digital companion-driven by wearable data, ecological momentary assessment, and adaptive feedback-can support relapse prevention and emotional self-regulation in high-vulnerability populations. We describe the conceptual design of the Digital-PA Loop, a system that interprets users’ daily patterns and delivers tailored interventions aligned with therapeutic goals. Ethical considerations, clinical integration pathways, and implementation strategies are discussed in detail. While the app remains under development, a pilot study involving patients at VillaRamadas is planned to assess feasibility, usability, and early signals of effectiveness. This proposal seeks to foster interdisciplinary collaboration at the intersection of sport science, digital health, and psychotherapy, and sets the stage for a data-driven evolution in addiction rehabilitation.

Area 3 - Sport Performance and Support Technology

Full Papers

Paper Nr:	19
Title:	Play Evaluation Based on Predicting the Outcome of Back-Row Attacks in Volleyball
Authors:	Hikaru Yoshihara, Ning Ding and Keisuke Fujii
Abstract:	In volleyball, statistical analysis based on data aggregation at the team or match levels has developed, and its use for player performance evaluation and tactical analysis has expanded. However, there has been limited discussion on the quantitative evaluation of how individual plays affect rally outcomes. To address this issue, a model that predicts rally outcomes under specific conditions using player location data is useful. This study aims to evaluate plays based on a prediction model, focusing on the first transition following a back-row attack. We extracted 103 target scenes from game footage recorded from behind the end line and manually created tracking data for six players per team. Using this dataset, we trained an XGBoost model to predict the future probability of scoring and the probability of blocking by two or more opponents in each game state (receive, toss, attack). To quantify play evaluation, we propose the Valuating Volleyball States by Estimating Probabilities (V2SEP), which expresses play evaluation values in each state based on the prediction model, weighting them according to the percentage of points scored when a player is blocked. To verify the validity of the prediction model used in V2SEP, we assessed F1 scores and SHAP values for each state. The results indicate that the predictions were reasonably accurate and reflected not only the contributions of directly involved players but also those of other players affecting scoring and block induction. Furthermore, the play evaluation metrics demonstrate expected trends whereas some scenes show the limitations, suggesting that V2SEP may be useful for play evaluation in volleyball.

Paper Nr:	31
Title:	A Generalized Valuation Method for Team Defense by Estimating Probabilities in Football Games
Authors:	Rikuhei Umemoto, Kazushi Tsutsui and Keisuke Fujii
Abstract:	Analyzing team defense in soccer is challenging due to limited labeled data. Some previous methods for evaluating soccer defenses were based on the prediction of defensive events using the locations of all players and the ball. However, they did not consider the importance of multiple events and assumed perfect observation of all 22 players, which is not open-source, with a larger amount for learning the classifier. In this paper, we propose a generalized valuation method for defensive teams by score-scaling the predicted probabilities of events, including gaining possession of the ball and being attacked. Our method can be applied to the open-source location data of all players in frames from broadcast video of events, such as football games from Euro 2020, by investigating the effect of the number of players on event prediction performance. Our validation results using Euro 2020 data show that event prediction accuracy can be maintained with a limited number of player features for scoring, conceding, gaining the ball, and effective attacks. Additionally, our defensive metric effectively explains the defensive characteristics and strengths of the top four teams in the tournament, while also highlighting the reasons why some teams received poor defensive evaluations. Our approach offers a practical way to analyze and evaluate team defenses even with self-recorded or broadcast videos.

Paper Nr:	57
Title:	Undulatory Underwater Swimming Performance and Kinematics: International- vs National-Level Female Swimmers
Authors:	Jesús J. Ruiz-Navarro, Adrián Febles-Castro, Óscar López-Belmonte, Francisco Cuenca-Fernández and Raúl Arellano
Abstract:	This study aimed to compare undulatory underwater swimming (UUS) performance and kinematics between international- and national-level female swimmers. Seven international level female swimmers (18.9 ± 3.4 years; 67.2 ± 4.4 kg of body mass; 175 ± 4.8 cm of body height; and 804 ± 35 World Aquatics points) and seven national level female swimmers (17.6 ± 1.7 years; 57.4 ± 4.1 kg of body mass; 166 ± 4.9 cm of body height; and 662 ± 65 World Aquatics points) performed three maximal-effort 15 m UUS trials. Seven body landmarks were auto-digitalized during UUS by a pre-trained neural network and 21 kinematic variables were calculated. The results showed no statistically significant differences across the variables analysed (p > 0.05); however, mean and minimum UUS velocities showed a clear trend toward better performance in internationallevel swimmers (p = 0.055–0.057; d = 0.91–0.92). In conclusion, there was a tendency for superior UUS performance among international-level swimmers compared to national-level swimmer. Furthermore, variations in UUS technique appear to be more strongly influenced by individual physical and anatomical characteristics than by performance level alone.

Short Papers

Paper Nr:	37
Title:	Wearable System for Measuring Impact Force and Response Time in Taekwondo Using Piezoresistive Sensors
Authors:	Elías Nicolás Chato-Cantos, Juan José Rivera-Zamora, Ana Cecilia Villa-Parra, Pablo Cevallos-Larrea and Mario Alvarez-Alvarez
Abstract:	This paper presents the design and preliminary validation of a wearable prototype integrated into a Taekwondo chest protector. The system combines piezoresistive force sensors and visual LED stimuli to measure two critical performance parameters: impact force and response time. Unlike traditional measurement systems that are limited to laboratory settings, the proposed device enables real-time assessment in realistic training environments. The sensors were calibrated using a progressive load method, and a linear model without intercept was selected to ensure proportional accuracy. Six athletes participated in the experimental protocol, executing the Bandal Chagui technique under sequential and random visual conditions. Results showed a decrease in impact force and an increase in response time under the random condition, suggesting that cognitive load affects technical performance. The system proved to be portable, low-cost, and suitable for integration into regular training routines.

Paper Nr:	40
Title:	Recovery and Readiness Monitoring Using Wearable Technology in Young Triathlon Athletes
Authors:	Ine De Bot, Jasper Gielen and Jean-Marie Aerts
Abstract:	This study explored the concept of ‘readiness to perform’ by monitoring twelve youth triathletes (under 23 (U23) and 19 (U19) years old) over three months using the Oura Ring. Physiological data from the wearable were analyzed for all participants; subjective assessments of training intensity (Rating of Perceived Exertion (RPE)) and recovery (Total Quality Recovery (TQR) questionnaire) were conducted only in the U23 subgroup. Stepwise linear regression was used to describe five (Balance) Scores contributing to the Readiness Score (RS). Subsequently, given the limited transparency of Oura’s algorithm, the RS was modeled using three approaches: through (1) its real contributors (RMSE = 3.18, R² = 0.71), (2) approximated contributors via regression and three additional contributors (RMSE = 4.09, R² = 0.52), and (3) directly measured variables with RPE and TQR scores (RMSE = 4.88, R² = 0.29). Individual-level analysis was prioritized, though a general model for describing the RS was also developed (RMSE = 3.48, R² = 0.60). Sleep emerged as the primary contributor to readiness, followed by physical activity and resting heart rate.

Paper Nr:	45
Title:	A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education
Authors:	Nelson Bilber Rodrigues, Rui Jorge Ramos, Mafalda Castro, Nuno Jesus, Pedro Guedes, Miguel Soares Ferreira, Rafael Silva and Lino Oliveira
Abstract:	Integrating Internet of Things (IoT) technologies into physical education (PE) presents opportunities for improving the methodologies for collecting, analysing, and managing student performance data. However, it also introduces technical challenges, particularly related to the real-time handling and protection of sensitive data in dynamic training environments. This paper presents a comprehensive solution outline based on a private local network architecture that supports scalable sensor data processing, real-time database integration, and mobile application interfaces. The proposed distributed system ensures data integrity, low-latency communication, and secure access while enabling educators to monitor student performance in real-time and review historical data. The system supports more personalised, data-driven training strategies by providing actionable insights for sports education.

Paper Nr:	47
Title:	Injury Risk Assessment in Women’s Football: Are We on the Right Path to Reduce Injury Risk? A Preliminary Systematic Scoping Review
Authors:	Asier Intxaurbe Gorostiza, Ibai Garcia-Tabar and Igor Setuain
Abstract:	Elite women’s football has experienced exponential growth, accompanied by an increasing in physical demands and a simultaneous shift in injury patterns. This scoping systematic review synthesised the methodologies employed in multicomponent screening tests (MCST) designed to mitigate injury risk among professional and semi-professional female players. Searches were conducted in PubMed/MEDLINE, Scopus and Web of Science (from inception in July 2024 to June 2025). Included studies were quantitative studies conducted on female soccer players. Main outcomes were functional screening profiles. Screening, data extraction, and quality assessment (Quality Assessment Tool for Quantitative Studies and the Oxford Levels of Evidence scales) were performed. Methods and results were reported according to PRISMA guidelines. The search yielded 4742 articles, of which 8 were included. Overall methodological quality of the studies was strong, with a moderate level of evidence. MCST protocols assessed mobility, lower-limb strength, core stability, jump mechanics and sprint mechanics, yet displayed considerable heterogeneity in instrumentation, metrics and cut off values. Despite growing interest, no universal protocol or consensus has yet emerged, hampering cross-study comparison and the formulation of robust injury preventive strategies for elite women’s football.

Paper Nr:	59
Title:	The Relationship Between Maximum Ball Throwing Speed and Shooting Accuracy and Expert Assessment of Basic Shot Technique in Handball
Authors:	Igor Gruić and Tomislav Jonjić
Abstract:	The aim of this study was to examine the relationship between anthropometric characteristics, handgrip strength, ball exit speed, shooting accuracy, and expert assessment of the basic handball shot technique. The research was conducted on a sample of 88 first-year students at the Faculty of Kinesiology. Measurements included ball speed (radar-based), shooting accuracy (goal segment scoring system), handgrip strength (dynamometer), and expert technique assessment (video evaluation). The results showed a significant correlation between expert assessment and ball speed, as well as a moderate correlation between certain anthropometric measures and ball velocity. No significant relationship was found between handgrip strength and shooting accuracy. In conclusion, technical execution and body characteristics play an important role in generating a powerful and effective handball shot, while shooting accuracy appears to rely on additional specific qualities beyond physical abilities alone.

Paper Nr:	22
Title:	Boutlength: Towards Identification and Presentation of Bout Lengths Using Physical Activity Counts Data
Authors:	Muhammad Asad Ullah Khan, Francesca Gallè and Giuliana Valerio
Abstract:	Boutlength is an open program developed in a spreadsheet environment to identify and display activity bout lengths derived from accelerometer count data. The program classifies counts into an active or inactive level based on user-defined cut-off thresholds. Using a bout qualification criterion specified by a minimum length, the program identifies, prints, sorts, and computes basic statistics on resulting valid bout string. The program was tested on a sample of actual database using cut-off values (Sedentary: < 100 counts; Activity: > 2019 counts) and bout qualification criteria (Sedentary: ≥ 5 minutes; Activity: ≥ 2 minutes). Output of the program is a display of identified Sedentary or Activity bouts depending on the application. Program’s parameters are modifiable, and the script was designed in a spreadsheet software. Boutlength is still in development and yet it may be an interesting resource for researchers analysing bout-based physical activity pattern measures.

Paper Nr:	27
Title:	Performance Evaluation of Penché Rotation in Rhythmic Gymnastics Using Statistical Parametric Mapping
Authors:	Bat-Otgon Batsuren, Batbayar Khuyagbaatar, Enkhsaikhan Gombojav, Battsetseg Gonchoo, Bayarjargal Ulziikhutag, Altantsetseg Tseveg and Yeruulbat Galbadrakh
Abstract:	In rhythmic gymnastics (RG), maintaining balances is essential for the successful execution of routines. In coaching practice, objective tools for assessing balance during routine execution are essential. Kinematic movement patterns have been analyzed using statistical parametric mapping (SPM), which evaluates movement and improves the understanding of tasks. This study examined three-dimensional (3D) lower extremity joint angles during static balance exercises and penché rotation in RG, then evaluated performance in penché rotation with SPM. The results showed a significant difference in the joint angles of the supporting leg during the initiation of rotation, while this difference tended to persist throughout the entire rotation for the lifted leg. This may indicate which specific joint motions do not align with threshold values in the movement patterns of the static balance test, which can be interpreted as a performance issue in dynamic rotation. This underscores SPM as a valuable tool for evaluating performance during rotation techniques in RG.

Paper Nr:	41
Title:	Variation of Kinematic and Dynamic Parameters with the Use of Minimalist Shoes in the Entry Phase of the Hammer Throw
Authors:	Gian Mario Castaldi, Sebastiano Conci, Andrea Amodio, Filippo Goi, Silvia Camboni, Alessandro Di Gregorio and Valentina Camomilla
Abstract:	This study investigates the biomechanical implications of footwear choice on the start phase of the hammer throw, aiming to determine whether minimalist footwear can be integrated into training without negatively affecting technical execution. A cohort of six trained hammer throwers performed the initial three rotations (start phase) under two different footwear conditions: standard World Athletics-approved throwing shoes and minimalist Fivefingers shoes by Vibram. Participants were divided between two motion capture laboratories for geographical reasons while ensuring consistent methodological application across environments. The objective was to assess whether the minimalist footwear-known to enhance activation of intrinsic foot musculature-alters key technical elements of the throwing motion. Analysis was structured around five biomechanically relevant instants within the entry phase, before the start of the turns, enabling comparison across footwear conditions of five parameters relative to the hammer head and the right foot motion, obtained using stereophotogrammetry (tangential hammer velocity, right joint movements and ground reaction forces. Preliminary results aim to determine whether the use of minimalist footwear brings an advantage to the throwers in the entry phase of throwing and whether it can be a useful tool for use in training.