Application of automated intelligent sensing technology in biomechanical characteristic analysis

Abstract

To clarify the specific effects of sports biomechanical regulation on improving technical performance and preventing sports injuries in rugby and soccer players, this study employed a randomized controlled trial design. A total of 48 rugby players (24 males and 24 females) and 40 soccer players (20 males and 20 females) were enrolled and randomly assigned to an experimental group (receiving an 8-week personalized biomechanical intervention) and a control group (undergoing conventional training) by gender stratification. The intervention protocol was developed and dynamically optimized based on high-precision kinematic and kinetic data collected by a multi-channel synchronous sensor system: 8 infrared high-speed motion capture cameras (sampling frequency: 200 Hz) were used to obtain the 3D motion trajectories of the athletes’ lower limb joints; 16-channel wireless surface electromyography (sEMG) sensors (sampling frequency: 1500 Hz) were applied to monitor the activation timing and amplitude of the quadriceps femoris, hamstrings, and lateral ankle muscle groups; and a 3D force platform (1000 Hz) was utilized to synchronously record ground reaction forces and lower limb joint torque data. The core of the intervention focused on optimizing the angular and torque parameters of the hip, knee, and ankle joints. Statistical analyses were performed using repeated-measures analysis of variance (ANOVA) and independent samples t-tests. The intraclass correlation coefficient (ICC) was used to assess the reliability of coaches’ scores, with ICC values ranging from 0.89 to 0.93. For technical improvement, the average skill scores of male/female rugby players in the experimental group increased from 57.83 ± 5.31 / 55.33 ± 2.87 to 68.42 ± 5.35 / 65.33 ± 3.67 (all P < 0.001); those of male/female soccer players rose from 41.85 ± 5.72 / 49.70 ± 5.13 to 58.75 ± 5.28 / 74.35 ± 6.89 (all P < 0.001). Time-frequency analysis based on sEMG sensors revealed that the percentage of myoelectric activity in key muscle groups was significantly higher in the experimental group than in the control group (e.g., a 7% increase in males and 3% in females for the lateral ankle muscle group of rugby players, P < 0.05), indicating optimized muscle activation efficiency. Regarding injury prevention, after the intervention, the hip frontal torque of female rugby players in the experimental group decreased to 8.73 ± 0.32 N·m (control group: 9.14 ± 0.41 N·m, P = 0.03), and the ankle coronal torque of male soccer players decreased by 8.6% (P = 0.02). The injury incidence during the intervention period was significantly lower in the experimental group (3.1%) than in the control group (15.6%, P = 0.01). This study confirms that multi-sensor fusion-based sports biomechanical intervention can simultaneously improve technical performance and reduce injury risk in rugby and soccer players by optimizing lower limb joint mechanical parameters and muscle activation patterns. It provides quantitative sensor data support and a precision intervention basis from an engineering perspective for the formulation of gender-specific specialized training programs.

Keywords

sports biomechanics / technical level / sports injury / mechanical analysis

References

[1] Yeadon M and Challis JH (1994) The future of performance—related sports biomechanics research. J Sports Sci 12: 3-32. https://doi.org/10.1080/02640419408732156

[2] Khera P and Kumar N (2020) Role of machine learning in gait analysis: a review. J Med Eng Technol 44: 441-467. https://doi.org/10.1080/03091902.2020.1822940

[3] Valueva MV, Nagornov N, Lyakhov PA, et al. (2020) Application of the residue number system to reduce hardware costs of the convolutional neural network implementation. Math Comput Simul 177: 232-243. https://doi.org/10.1016/j.matcom.2020.04.031

[4] Acikkar M, Akay MF, Ozgunen KT, et al. (2009) Support vector machines for aerobic fitness prediction of athletes. Expert Syst Appl 36: 3596-3602. https://doi.org/10.1016/j.eswa.2008.02.002

[5] Garcia G, Bucheli R, Castillo JT, et al. (2025) Biomechanical and neuromuscular differences between professional and varsity soccer players during countermovement and approach jumps. Plos One 20: e0336672. https://doi.org/10.1371/journal.pone.0336675

[6] Sikora O, Lehnert M, Hanzlíková I, et al. (2023) The impact of a novel neuromuscular training program on leg stiffness, reactive strength, and landing biomechanics in amateur female rugby players. Appl Sci 13: 1979. https://doi.org/10.3390/app13031979

[7] Taylor JB, Ford KR, Schmitz RJ, et al. (2018) Sport—specific biomechanical responses to an ACL injury prevention programme: a randomised controlled trial. J Sports Sci 36: 2492-2501. https://doi.org/10.1080/02640414.2018.1465723

[8] Field AC, Harper LD, Aldous JWF, et al. (2023) A commentary on soccer match—play simulations for applied research and practice. Sci Med Football 7: 93-105. https://doi.org/10.1080/24733938.2022.2070268

[9] Sannicandro I (2021) Small—side games and size pitch in elite female soccer players: A short narrative review. J Hum Sport Exerc 16: S361-S369. https://doi.org/10.14198/jhse.2021.16.Proc2.20

[10] Zhang JY, Li QY, Li JC, et al. (2026) High—sensitivity flexible triboelectric nanogenerator sensor based on recycled PA66 for the monitoring of soccer player lower limb training. Mater Res Bull 194: 113774. https://doi.org/10.1016/j.materresbull.2025.113774

[11] Schrapf N, Hassan A, Wiesmeyr S, et al. (2022) An artificial neural network predicts setter’s setting behavior in volleyball similar or better than experts. IFAC—Papers OnLine 55: 612-617. https://doi.org/10.1016/j.ifacol.2022.09.163

[12] Wang CS, Wang LH, Kuo LC, et al. (2017) Comparison of breast motion at different levels of support during physical activity. J Hum Sport Exerc 12: 1256-1264. https://doi.org/10.14198/jhse.2017.124.12

[13] Krizhevsky A, Sutskever I, Hinton GE (2017) Imagenet classification with deep convolutional neural networks. Commun ACM 60: 84-90. https://doi.org/10.1145/3065386

[14] Bridgman C, Scurr J, White J, et al. (2010) Three—dimensional kinematics of the breast during a two—step star jump. J Appl Biomech 26: 465-472. https://doi.org/10.1123/jab.26.4.465

[15] Knudson D and Wallace B (2021) Student perceptions of low—tech active learning and mastery of introductory biomechanics concepts. Sport Biomech 20: 458-468. https://doi.org/10.1080/14763141.2019.1570322

[16] McGhee DE and Steele JR (2010) Optimising breast support in female patients through correct bra fit. a cross—sectional study. J Sci Med Sport 13: 568-572. https://doi.org/10.1016/j.jsams.2010.03.003

[17] Keogh JW, Moro C, Knudson D (2024) Promoting learning of biomechanical concepts with game—based activities. Sport Biomech 23: 253-261. https://doi.org/10.1080/14763141.2020.1845470

[18] Hamill J, Knutzen KM, Derrick TR (2021) Biomechanics: 40 years on. Kinesiology Rev 10: 228-237. https://doi.org/10.1123/kr.2021—0015

[19] Woods ML, Karp GG, Feltz DL (2003) Positions in kinesiology and physical education at the college or university level. Quest 55: 30-50. https://doi.org/10.1080/00336297.2003.10491787

[20] Ceylan B, Simenko J, Balci SS (2022) Which performance tests best define the special judo fitness test classification in elite judo athletes? J Funct Morphol Kinesiol 7: 101. https://doi.org/10.3390/jfmk7040101

[21] Blanksby B, Nicholson L, Elliott B (2002) Swimming: biomechanical analysis of the grab, track and handle swimming starts: an intervention study. Sport Biomech 1: 11-24. https://doi.org/10.1080/14763140208522784