Development of a Physiotherapy Exercise Monitoring System with LSTM Auto-Encoder

• Luís Henrique C. Evangelista UEA
• Carlos Maurício S. Figueiredo UEA
• Elloá B. Guedes UEA

DOI: https://doi.org/10.5753/sbcup.2024.2662

Abstract

This work aims to contribute to the intelligent monitoring of physiotherapy exercises for telerehabilitation purposes. We propose a Computer Vision solution based on Deep Learning with Auto-Encoders and LSTMs for classifying exercise videos according to the correctness of their execution. Integrating non-invasive pose landmark estimation and anomaly detection techniques, the proposed solution contributes to the monitoring and feedback of remote patients, which can positively collaborate with treatment adherence.

References

Boukhennoufa, I., Zhai, X., Utti, V., Jackson, J., and McDonald-Maier, K. D. (2022). Wearable sensors and machine learning in post-stroke rehabilitation assessment: A systematic review. Biomedical Signal Processing and Control, 71:103197.

Brink, H., Richards, J. W., and Fetherolf, M. (2017). Real-World Machine Learning. Manning Publications, Estados Unidos.

Dorado, J., del Toro, X., Santofimia, M. J., Parreño, A., Cantarero, R., Rubio, A., and Lopez, J. C. (2019). A computer-vision-based system for at-home rheumatoid arthritis rehabilitation. International Journal of Distributed Sensor Networks, 15(9):1550147719875649.

Google (2023). Pose landmarker solution. Acesso em 24 de junho de 2023: [link].

Hellsten, T., Karlsson, J., Shamsuzzaman, M., and Pulkkis, G. (2021). The potential of computer vision-based marker-less human motion analysis for rehabilitation. Rehabilitation Process and Outcome, 10:11795727211022330.

Latreche, A., Kelaiaia, R., Chemori, A., and Kerboua, A. (2023). Reliability and validity analysis of mediapipe-based measurement system for some human rehabilitation motions. Measurement, 214:112826.

Lugaresi, C., Tang, J., Nash, H., McClanahan, C., Uboweja, E., Hays, M., Zhang, F., Chang, C.-L., Yong, M. G., Lee, J., et al. (2019). Mediapipe: A framework for building perception pipelines. arXiv preprint arXiv:1906.08172. Acesso em 26 de junho de 2023: [link].

OMS (2024). Rehabilitation. Disponível em [link]. Acesso em 24 de maio de 2024.

Prvu Bettger, J. and Resnik, L. J. (2020). Telerehabilitation in the age of COVID-19: An opportunity for learning health system research. Phys. Ther., 100(11):1913–1916.

Sigal, L. (2021). Human Pose Estimation, pages 573–592. Springer International Publishing, Cham.

Subtil, M. M. L., Goes, D. C., Gomes, T. C., and Souza, M. L. d. (2011). O relacionamento interpessoal e a adesão na fisioterapia. Fisioterapia em Movimento, 24(4):745–753.

ToInfinityAI (2023). Infiniterep dataset. [link].

Vakanski, A., Jun, H.-p., Paul, D., and Baker, R. (2018). A data set of human body movements for physical rehabilitation exercises. Data, 3(1):2.