Peningkatan Akurasi Deteksi Anomali Biomekanika melalui Model Probabilistik Berbasis Fusi Vision

Ahmad Indra Harahap

doi:10.60076/indotech.v4i1.2039

Authors

Ahmad Indra Harahap Universitas Negeri Medan

DOI:

https://doi.org/10.60076/indotech.v4i1.2039

Keywords:

Anomali Biomekanik, Fusi Visi, Ambang Beban, Validasi Mediapipe, Umpan Balik Real-time

Abstract

Risiko cedera lumbal dan ACL pada atlet powerlifting akibat deviasi teknik mikro menuntut adanya instrumen pemantau biomekanika yang presisi namun ekonomis. Penelitian ini bertujuan mengonstruksi Model Probabilistik Deteksi Anomali yang memfusikan algoritma Computer Vision dengan Logika Geometri Euclidean sebagai detektor gerak waktu nyata. Metode penelitian melibatkan fusi estimasi pose MediaPipe dengan protokol validasi biomedis untuk menetapkan ambang batas aman pada gerakan Squat, Bench Press, dan Deadlift. Hasil penelitian menunjukkan bahwa model mampu mendeteksi deviasi teknik dengan tingkat akurasi sebesar 93% dan latensi komputasi di bawah 50 milidetik, yang memungkinkan umpan balik visual korektif secara instan. Verifikasi klinis oleh pakar biomedis mengonfirmasi bahwa landmark digital memiliki korelasi tinggi dengan standar teknik baku. Simpulannya, integrasi vision-probabilistik ini menyediakan solusi pemantauan biomekanika mandiri yang efektif untuk menekan prevalensi cedera pada atlet tanpa bergantung pada perangkat laboratorium yang mahal

Downloads

Download data is not yet available.

References

E. Strömbäck, U. Aasa, K. Gilenstam, and L. Berglund, "Prevalence and incidence of injuries in powerlifting: A systematic review," Orthopaedic Journal of Sports Medicine, vol. 6, no. 5, 2018.

L. Petrigna, A. Karsten, G. Marcolin, and A. Paoli, "A review of mechanical and physiological aspects of the Squat exercise," Journal of Human Kinetics, vol. 74, no. 1, pp. 67-82, 2020.

M. J. Escamilla, R. F. Escamilla, and J. E. Fleisig, "Biomechanics of the Deadlift exercises: A systematic review," International Journal of Sports Science & Coaching, 2022.

F. Sgrò, M. Lipoma, and T. Lovecchio, "Evaluating the validity of a low-cost motion analysis system for detecting upper body posture alterations," Journal of Human Sport and Exercise, vol. 15, no. 2, pp. 320-332, 2020.

A. I. Ma’ruf, "Computer Vision-based exercise technique monitoring," Indonesian Journal of Computing and Cybernetics Systems (IJCCS), 2021.

T. Needham, M. Naeem, and M. A. Azam, "A comparative study of OpenPose and MediaPipe for physical exercise monitoring," Sensors, vol. 22, no. 1, 2022.

K. Chen, Y. Zhang, and Z. Li, "Augmented Reality in sports training: A systematic review," Applied Sciences, vol. 13, no. 5, 2023.

A. M. M. Al-Saffar, H. Tao, and M. A. Talab, "Review of deep learning techniques for sport performance analysis," IEEE Access, vol. 9, pp. 28530-28545, 2021.

S. K. Yadav, A. Tiwari, H. M. Pandey, and S. A. Akbar, "A review of multimodal human activity recognition with special emphasis on deep learning," IEEE Access, vol. 9, 2021.

E. E. Cust, A. J. Sweeting, K. Ball, and S. Robertson, "Machine and deep learning for sport-specific movement recognition: A systematic review," Journal of Sports Sciences, vol. 37, no. 5, pp. 568-600, 2019.

R. G. D. Silva, "Real-time Squat posture correction system using Computer Vision," International Journal of Advanced Computer Science and Applications (IJACSA), vol. 13, no. 5, pp. 112-119, 2022.

J. H. Park and S. H. Lee, "Deep learning-based visual feedback system for weight training posture correction," Applied Sciences, vol. 11, no. 21, 2021.

V. Ramjit and K. Sabanayagam, "Postural detection and correction of Squats using deep learning," in 2021 IEEE International Conference on Computation System and Information Technology for Sustainable Solutions (CSITSS), 2021, pp. 1-6.

G. Ribeiro, A. H. Sousa, and J. P. Papa, "Human pose estimation for sports analytics: A survey," IEEE Transactions on Human-Machine Systems, vol. 53, no. 1, pp. 1-13, 2023.

Y. Kong, Y. Liu, and Y. Fu, "Kinematic descriptions of human motion: A survey," Computer Vision and Image Understanding, vol. 191, 2020.

C. Lugaresi et al., "MediaPipe: A framework for building perception pipelines," arXiv preprint arXiv:1906.08172, 2019.

Z. Cao, G. Hidalgo, T. Simon, S. E. Wei, and Y. Sheikh, "OpenPose: Realtime multi-person 2D pose estimation using Part Affinity Fields," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 172-186, 2021.