Construction of a Student Sports Risk Early Warning Model Based on Multi-Source Data Fusion

Miao Zhou1 , Huicong Lei1*
1 International Communication and General Education School, Guangxi College of Sports Education, Nanning, Guangxi, China
* Corresponding author: Huicong Lei. Email: 15578212585@163.com
Journal of Digital Frontier 2026, Vol. 1, No. 1, pp. 31-60
DOI: 10.67541/jdf2602
Received: 29 May 2026; Revised: 29 June 2026; Accepted: 10 July 2026; Published: 15 July 2026
Abstract

In response to the frequent occurrence of student sports injuries and the lack of real-time warning methods in campus sports scenarios, this paper proposes a student sports risk warning model based on multi-source data fusion. A multi-source acquisition system incorporating wearable physiological signals, video-based human postures, and environmental sensors is constructed, and an adaptive weighted early fusion strategy is proposed to dynamically adjust the contribution of each data source. At the feature extraction level, a multi-scale temporal convolutional network with a multi-head attention mechanism (MHA-MSTCN) and a residual contraction module is designed to capture multi-time-scale risk patterns and suppress motion noise. Furthermore, a dual-path heterogeneous fusion architecture is proposed, using graph attention networks to model joint coupling abnormalities in human motion chains, gated recurrent units to capture temporal drift of physiological parameters, and cross-branch cross-attention to achieve collaborative discrimination of spatial and temporal features. Unsupervised clustering of risk levels is achieved using K-means++ and the silhouette coefficient, and the warning trigger is completed by combining individualized dynamic thresholds and a multi-level delay-tolerant rule engine. Experimental results on a self-built dataset (120 subjects, 53.6 hours of sports data) show that the model has an accuracy of 92.3%, a recall rate of 89.7%, an AUC of 97.1%, a warning lead time of 2.31 seconds, a single-frame inference delay of 7.6 milliseconds, and is significantly superior to baseline models such as LSTM, CNN, and Transformer. Ablation experiments verify the effectiveness of each core module. This model provides a feasible technical solution for intelligent campus sports risk prevention and control.

Keywords
Multi-source data fusion Sports risk warning Time-series convolutional network Graph attention network Edge computing
References
  1. Al-Qahtani, M. A., Allajhar, M. A., Alzahrani, A. A., Asiri, M. A., Alsalem, A. F., Alshahrani, S. A., ... & Alsalem, A. F. F. (2023). Sports-related injuries in adolescent athletes: a systematic review. Cureus, 15(11). DOI: 10.7759/cureus.49392
  2. Zeng, Y., Ji, W., Shi, Y., Liu, W., & Ji, W. (2025). Sports injuries in elite football players: Classification, prevention, and treatment strategies update. Frontiers in Sports and Active Living, 7, 1643789. DOI: 10.3389/fspor.2025.1643789
  3. Nair, M. J., & Swarna Nantha, Y. (2025). Common Sports, Muscular, and Ligamentous Injuries. In Screen, Diagnose, Manage and Monitor: A GP Primer to Common Clinical Conditions (pp. 1275-1283). Singapore: Springer Nature Singapore. DOI: 10.1007/978-981-97-7640-5_146
  4. Sixt, J. C. (2026). High School Baseball Coaches Understanding and Management of Potential Mild Traumatic Brain Injuries in Catchers (Doctoral dissertation, Grand Canyon University).
  5. Porsanger, L. (2023). Risk and safety management in physical education: teachers' knowledge. Physical Education and Sport Pedagogy, 28(1), 16-28. DOI: 10.1080/17408989.2021.1934663
  6. Zou, X., Wang, L., & Han, R. (2025). Embedded Cyber-Physical Systems for Monitoring and Optimization of Sports Facilities Using Consumer IoT (CIoT) and Edge Computing. IEEE Transactions on Consumer Electronics, 71(2), 4838-4848. DOI: 10.1109/tce.2025.3535162
  7. Jia, Y. (2026). Integrating Heart Rate Variability and Exercise Load for Early Warning of Overtraining Syndrome. International Journal of Computer Information Systems and Industrial Management Applications, 18, 12-12. DOI: 10.70917/ijcisim-2026-1827
  8. Urbina, R. (2022). Utilizing Heart Rate as an Exercise Intensity Training Method to Determine Overtraining for Collegiate Long-Distance Runners. SACAD: Scholarly Activities, 2022(2022), 12. DOI: 10.58809/gqki2597
  9. Shariatzadeh, M., Hafshejani, E. H., Mitchell, C. J., Chiao, M., & Grecov, D. (2023). Predicting muscle fatigue during dynamic contractions using wavelet analysis of surface electromyography signal. Biocybernetics and Biomedical Engineering, 43(2), 428-441. DOI: 10.1016/j.bbe.2023.04.002
  10. Sarvestan, J., & Fakhraei Rad, N. (2025). Lower Limb Joint Coordination and Coordination Variability During Landing: A Scoping Review. Applied Sciences, 15(9), 5118. DOI: 10.3390/app15095118
  11. Khosravi, M. R. (2025). Sports Injuries: Biomechanical Data Analysis and Prevention. Advanced Journal of Management, Humanity and Social Science, 1(9), 559-578. DOI: 10.23977/JHMS.2025.060106
  12. Moreira, R., Fialho, R., Teles, A. S., Bordalo, V., Vasconcelos, S. S., de Morais Gouveia, G. P., ... & Teixeira, S. (2022). A computer vision-based mobile tool for assessing human posture: A validation study. Computer Methods and Programs in Biomedicine, 214, 106565. DOI: 10.1016/j.cmpb.2021.106565
  13. Wang, X. (2024). Biomechanical Analysis on the In-situ Three-point Jump Shooting Technique in Basketball. In Sustainable Materials Processing and Manufacturing (pp. 1-6). CRC Press. DOI: 10.1201/9781003538646-1
  14. Yan, W., Jiang, X., & Liu, P. (2023). A review of basketball shooting analysis based on artificial intelligence. IEEE Access, 11, 87344-87365. DOI: 10.1109/access.2023.3304631
  15. Mason, R., Pearson, L. T., Barry, G., Young, F., Lennon, O., Godfrey, A., & Stuart, S. (2023). Wearables for running gait analysis: A systematic review. Sports Medicine, 53(1), 241-268. DOI: 10.1007/s40279-022-01760-6
  16. Man, K., & Chahl, J. (2022). A review of synthetic image data and its use in computer vision. Journal of Imaging, 8(11), 310. DOI: 10.3390/jimaging8110310
  17. Jaiswal, D., & Kumar, P. (2022). A survey on parallel computing for traditional computer vision. Concurrency and Computation: Practice and Experience, 34(4), e6638. DOI: 10.1002/cpe.6638
  18. Oyeleye, M., Chen, T., Titarenko, S., & Antoniou, G. (2022). A predictive analysis of heart rates using machine learning techniques. International Journal of Environmental Research and Public Health, 19(4), 2417. DOI: 10.3390/ijerph19042417
  19. Ghafoori, M., Clevenger, C., Abdallah, M., & Rens, K. (2023). Heart rate modeling and prediction of construction workers based on physical activity using deep learning. Automation in Construction, 155, 105077. DOI: 10.1016/j.autcon.2023.105077
  20. Gu, M., Zhang, Y., Wen, Y., Ai, G., Zhang, H., Wang, P., & Wang, G. (2023). A lightweight convolutional neural network hardware implementation for wearable heart rate anomaly detection. Computers in Biology and Medicine, 155, 106623. DOI: 10.1016/j.compbiomed.2023.106623
  21. Canessa-Pollard, V., Anikin, A., & Reby, D. (2025). Chants across seven traditions share acoustic traits that enhance subjective relaxation. Proceedings of the National Academy of Sciences, 122(35), e2506480122. DOI: 10.1073/pnas.2506480122
  22. Yunardi, R. T., Sardjono, T. A., & Mardiyanto, R. (2025). Enhancing Surveillance Vision-Based Human Action Recognition Using Skeleton Joint Swing and Angle Feature and Modified AlexNet-LSTM. International Journal of Intelligent Engineering & Systems, 18(1). DOI: 10.22266/ijies2025.0229.53
  23. Chen, H., & Fan, R. (2025). Improved convolutional neural network for precise exercise posture recognition and intelligent health indicator prediction. Scientific Reports, 15(1), 21309. DOI: 10.1038/s41598-025-01854-x
  24. Alif, F. K. (2026). Extending peaks-over-threshold methods to public health: modified threshold selection and scenario-based risk estimation. Health Services and Outcomes Research Methodology, 1-25. DOI: 10.21203/rs.3.rs-7553933/v1
  25. Liu, P. L., & Chang, C. C. (2022). Simple method integrating OpenPose and RGB-D camera for identifying 3D body landmark locations in various postures. International Journal of Industrial Ergonomics, 91, 103354. DOI: 10.1016/j.ergon.2022.103354
  26. Balakrishnan, K., Dhanalakshmi, R., & Gopalakrishnan, R. (2023). Clock synchronization in industrial Internet of Things and potential works in precision time protocol: Review, challenges and future directions. International Journal of Cognitive Computing in Engineering, 4, 205-219. DOI: 10.1016/j.ijcce.2023.06.001
  27. Xie, Q., Tao, G., Xie, C., & Wen, Z. (2022). Abnormal data detection based on adaptive sliding window and weighted multiscale local outlier factor for machinery health monitoring. IEEE Transactions on Industrial Electronics, 70(11), 11725-11734. DOI: 10.1109/tie.2022.3231279
  28. Zhao, F., Zhang, C., & Geng, B. (2024). Deep multimodal data fusion. ACM Computing Surveys, 56(9), 1-36. DOI: 10.1145/3649447
  29. Pant, A., & Kumar, A. (2024). Hanning FIR window filtering analysis for EEG signals. Biomedical Analysis, 1(2), 111-123. DOI: 10.1016/j.bioana.2024.05.003
  30. Ayodeji, A., Wang, Z., Wang, W., Qin, W., Yang, C., Xu, S., & Liu, X. (2022). Causal augmented ConvNet: A temporal memory dilated convolution model for long-sequence time series prediction. ISA Transactions, 123, 200-217. DOI: 10.1016/j.isatra.2021.05.026
  31. Giuste, F. O., He, L., Lais, P., Shi, W., Zhu, Y., Hornback, A., ... & Wang, M. D. (2023). Early and fair COVID-19 outcome risk assessment using robust feature selection. Scientific Reports, 13(1), 18981. DOI: 10.1038/s41598-023-36175-4
  32. Vrahatis, A. G., Lazaros, K., & Kotsiantis, S. (2024). Graph attention networks: a comprehensive review of methods and applications. Future Internet, 16(9), 318. DOI: 10.3390/fi16090318
  33. Zhang, Q., & Qian, J. (2025). Identification and Temporal distribution of typical rainfall types based on K-Means++ Clustering and probability distribution analysis. Hydrology, 12(4), 88. DOI: 10.3390/hydrology12040088