Advanced biomechanical analytics: Wearable technologies for precision health monitoring in sports performance

Digital Health, Journal Year: 2024, Volume and Issue: 10

Published: Jan. 1, 2024

Objective This study investigated the impact of wearable technologies, particularly advanced biomechanical analytics and machine learning, on sports performance monitoring intervention strategies within realm physiotherapy. The primary aims were to evaluate key metrics, individual athlete variations efficacy learning-driven adaptive interventions. Methods research employed an observational cross-sectional design, focusing collection analysis real-world data from athletes engaged in A representative sample Bahawalpur participated, utilizing Dring Stadium as venue. Wearable devices, including inertial sensors (MPU6050, MPU9250), electromyography (EMG) (MyoWare Muscle Sensor), pressure (FlexiForce sensor) haptic feedback sensors, strategically chosen for their ability capture diverse parameters. Results Key such heart rate (mean: 76.5 bpm, SD: 3.2, min: 72, max: 80), joint angles 112.3 degrees, 6.8, 105, 120), muscle activation 43.2%, 4.5, 38, 48) stress strain features [112.3 ], [6.5 ]), analyzed presented summary tables. Individual analyses highlighted emphasizing need personalized strategies. technologies athletic was quantified through a comparison metrics recorded with without sensors. consistently demonstrated improvements monitored parameters, affirming significance technologies. Conclusions suggests that when combined can enhance Real-time allows precise adjustments, demonstrating potential

Language: Английский

---

An Efficient Approach to Sports Rehabilitation and Outcome Prediction Using RNN-LSTM

Mobile Networks and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: June 12, 2024

Language: Английский

---

Cloud-to-Thing continuum-based sports monitoring system using machine learning and deep learning model

PeerJ Computer Science, Journal Year: 2025, Volume and Issue: 11, P. e2539 - e2539

Published: Feb. 10, 2025

Sports monitoring and analysis have seen significant advancements by integrating cloud computing continuum paradigms facilitated machine learning deep techniques. This study presents a novel approach for sports monitoring, specifically focusing on basketball, that seamlessly transitions from traditional cloud-based architectures to paradigm, enabling real-time insights into player performance team dynamics. Leveraging algorithms, our framework offers enhanced capabilities tracking, action recognition, evaluation in various scenarios. The proposed Cloud-to-Thing continuum-based system utilizes advanced techniques such as Improved Mask R-CNN pose estimation hybrid metaheuristic algorithm combined with generative adversarial network (GAN) classification. Our significantly improves latency accuracy, reducing 5.1 ms achieving an accuracy of 94.25%, which outperforms existing methods the literature. These results highlight system's ability provide real-time, precise, scalable immediate feedback time-sensitive applications. research has improved event analysis, contributing evaluation, strategies, informed tactical adjustments.

Language: Английский

---

Efficient Parkinson’s disease classification from dynamic foot pressure data: A combined approach of clustering and feature selection

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 105, P. 107654 - 107654

Published: Feb. 11, 2025

Language: Английский

---

Application of flexible sensor multimodal data fusion system based on artificial synapse and machine learning in athletic injury prevention and health monitoring

Discover Artificial Intelligence, Journal Year: 2025, Volume and Issue: 5(1)

Published: March 31, 2025

Language: Английский

---

Application of Wearable Insole Sensors in In-Place Running: Estimating Lower Limb Load Using Machine Learning

Biosensors, Journal Year: 2025, Volume and Issue: 15(2), P. 83 - 83

Published: Feb. 1, 2025

Musculoskeletal injuries induced by high-intensity and repetitive physical activities represent one of the primary health concerns in fields public fitness sports. injuries, often resulting from unscientific training practices, are particularly prevalent, with tibia being especially vulnerable to fatigue-related damage. Current tibial load monitoring methods rely mainly on laboratory equipment wearable devices, but datasets combining both sources limited due experimental complexities signal synchronization challenges. Moreover, wearable-based algorithms fail capture deep features, hindering early detection prevention fatigue injuries. In this study, we simultaneously collected data insole sensors during in-place running volunteers, creating a dataset named WearLab-Leg. Based dataset, developed machine learning model integrating Temporal Convolutional Network (TCN) Transformer modules estimate vertical ground reaction force (vGRF) bone (TBF) using pressure signals. Our model’s architecture effectively combines advantages local feature extraction global modeling, further introduces Weight-MSELoss function improve peak prediction performance. As result, achieved normalized root mean square error (NRMSE) 7.33% for vGRF 10.64% TBF prediction. proposed offer convenient solution biomechanical athletes patients, providing reliable technical support warnings fatigue-induced

Language: Английский

---

Optimization of Decision Support Technology for Offshore Oil Condition Monitoring with Carbon Neutrality as the Goal in the Enterprise Development Process

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(3), P. e0319858 - e0319858

Published: March 25, 2025

This study aims to explore the integration of Faster R-CNN (Region-based Convolutional Neural Network) algorithm from deep learning into MobileNet v2 architecture, within context enterprises aiming for carbon neutrality in their development process. The experiment develops a marine oil condition monitoring and classification model based on fusion algorithms. utilizes network extract rich feature information input images combines rapidly accurately generate candidate regions monitoring, followed by detailed these regions. performance is evaluated through experimental assessments. results demonstrate that average loss value proposed approximately 0.45. Moreover, recognition accuracy training testing sets reaches 90.51% 93.08%, respectively, while other algorithms remains below 90%. Thus, constructed this exhibits excellent terms accuracy, providing reliable technical support offshore contributing promotion sustainable utilization conservation resources.

Language: Английский

---

Research on Athlete Technical Action Scoring Model Based on Deep Learning

Learning and analytics in intelligent systems, Journal Year: 2025, Volume and Issue: unknown, P. 358 - 367

Published: Jan. 1, 2025

Language: Английский

---

From data to decision: Machine learning determination of aerobic and anaerobic thresholds in athletes

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(8), P. e0309427 - e0309427

Published: Aug. 29, 2024

Lactate analysis plays an important role in sports science and training decisions for optimising performance, endurance, overall success sports. Two parameters are widely used these goals: aerobic (AeT) anaerobic (AnT) thresholds. However, determining AeT proves more challenging than AnT threshold due to both physiological intricacies practical considerations. Thus, the aim of this study was determine thresholds using machine learning modelling (ML) compare ML-obtained results with parameters’ values determined conventional methods. ML seems be highly useful its ability handle complex, personalised data, identify nonlinear relationships, provide accurate predictions. The 183 CardioPulmonary Exercise Test (CPET) accompanied by lactate heart ratio analyses from amateur athletes were enrolled models following algorithms: Random Forest, XGBoost (Extreme Gradient Boosting), LightGBM (Light Boosting Machine) metrics: R 2 , mean absolute error (MAE), squared (MSE) root square (RMSE). regressors belong group ensemble algorithms that combine predictions multiple base improve performance counteract overfitting data. Based on evaluation metrics, give best predictions: AeT: Forest has value 0.645, MAE 4.630, MSE 44.450, RMSE 6.667; AnT: 0.803, highest among models, 3.439, lowest 20.953, 4.577. Outlined research experiments, a comprehensive review existing literature field, obtained suggest can trained make based individual athlete’s unique response exercise. Athletes exhibit significant variation their AT, capture differences, allowing tailored recommendations optimization.

Language: Английский

---

Infodemiology of Influenza-like Illness: Utilizing Google Trends’ Big Data for Epidemic Surveillance

Journal of Clinical Medicine, Journal Year: 2024, Volume and Issue: 13(7), P. 1946 - 1946

Published: March 27, 2024

Background: Influenza-like illness (ILI) encompasses symptoms similar to influenza, affecting population health. Surveillance, including Google Trends (GT), offers insights into epidemic patterns. Methods: This study used multiple regression models analyze the correlation between ILI incidents, GT keyword searches, and climate variables during influenza outbreaks. It compared predictive capabilities of time-series deep learning against emergency incidents. Results: The searches for “fever” “cough” were significantly associated with cases (p < 0.05). Temperature had a more substantial impact on incidence than humidity. Among tested models, ARIMA provided best power. Conclusions: data can forecast trends, aiding governmental decision making. is crucial predictor, excel in forecasting incidences.

Language: Английский

---