AI-Powered Real-Time Tele-Rehabilitation System Using Computer Vision for Orthopedic Patients

This methodology presents the design and implementation of an AI-powered computer vision–based physiotherapy rehabilitation system intended to support home and remote exercise monitoring. The proposed system utilizes real-time pose estimation through a standard webcam to detect body key points, compute joint angles, and evaluate exercise performance without the need for wearable sensors or invasive devices. By integrating machine learning–based movement classification with automated visual and audio feedback, the system provides objective posture correction and repetition tracking. Session data is recorded and analyzed to generate performance metrics and progress reports, enabling quantitative assessment of rehabilitation outcomes. The results demonstrate that camera-based motion analysis can offer a cost-effective, non-invasive, and scalable solution for enhancing accessibility and consistency in physiotherapy rehabilitation.In addition to improving accessibility, the system significantly reduces the dependency on continuous clinical supervision by enabling patients to perform exercises independently at home. The use of real-time feedback ensures that users are immediately guided to correct improper posture, thereby minimizing the risk of injury and improving exercise effectiveness. The system is designed with a user-friendly interface that allows individuals with minimal technical knowledge to operate it Furthermore, the integration of automated data logging ensures accurate tracking of patient performance over time, which can be useful for long-term rehabilitation planning. The ability to generate structured reports supports better communication between patients and physiotherapists. The system also promotes consistency in exercise routines by providing continuous monitoring and reminders.

Artificial Intelligence, Computer Vision, Physiotherapy Rehabilitation, Human Pose Estimation

1. Z. Cao, G. Hidalgo, T. Simon, S. Wei, and Y. Sheikh, “Open Pose: Real time 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. [Google Scholar] [Crossref]

2. C. Lugaresi et al., “Media Pipe: A Framework for Building Perception Pipelines,” ar Xiv preprint ar Xiv: 1906.08172, 2019. [Google Scholar] [Crossref]

3. J. Shotton et al., “Real-Time Human Pose Recognition in Parts from Single Depth Images,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011, pp. 1297–1304. [Google Scholar] [Crossref]

4. E. Olesh, S. Yakubenko, and T. Gollan, “Automated Assessment of Upper Limb Movement Impairment Using RGB-D Camera,” in Proceedings of the IEEE Engineering in Medicine and Biology Society (EMBC), 2014, pp. 606– 609. [Google Scholar] [Crossref]

5. X. Xu, R. W. McGorry, L. Chou, J. Lin, and C. Chang, “Accuracy of the Micros of t Kinect™ for Measuring Human Movement,” Gait & Posture, vol. 41, no. 2, pp. 557–562, 2015. [Google Scholar] [Crossref]

6. B. Bonnechère et al., “The Use of Commercial Video Games in Rehabilitation: A Systematic Review,” International Journal of Rehabilitation Research, vol. 41, no. 4, pp. 277–292, 2018. [Google Scholar] [Crossref]

7. S. Patel et al., “A Review of Wearable Sensors and Systems for Monitoring Motor Rehabilitation,” Journal of Neuro Engineering and Rehabilitation, vol. 9,no.1, p. 21, 2012. [Google Scholar] [Crossref]

8. Ł. Kidziński et al., “Deep Learning for Human Movement Analysis in Rehabilitation,” NPJ Digital Medicine, vol. 3, no. 1, pp. 1–8, 2020. [Google Scholar] [Crossref]

9. J. Wang, Y. Chen, S. Hao, X. Peng, and L. Hu, “Deep Learning for Sensor- Based Activity Recognition: A Survey,” Pattern Recognition Letters, vol. 119, pp.3–11,2019. [Google Scholar] [Crossref]

10. Y. Yang et al., “Vision-Based Rehabilitation Exercise Monitoring Using Deep Learning,” IEEE Access, vol. 9, pp. 103921–103934, 2021. [Google Scholar] [Crossref]

11. S. Li and T. Sato, “AI-Based Posture Correction System Using Computer Vision,” IEEE Access, vol. 8, pp. 163947–163958, 2020. [Google Scholar] [Crossref]

12. D. Dias and J. Cunha, “Wearable Health Devices—Vital Sign Monitoring Systems,” Sensors, vol. 18, no. 8, p. 2414, 2018. [Google Scholar] [Crossref]

13. R. Ramakrishnan et al., “A Deep Learning-Based Exercise Recognition System Using Wearable Sensors,” IEEE Sensors Journal, vol. 19, no. 19, pp. 8970–8979, 2019. [Google Scholar] [Crossref]

14. K. Chen et al., “Intelligent Home-Based Rehabilitation System Using Computer Vision and AI,” Healthcare Technology Letters, vol. 9, no. 2, pp.45– 52, 2022. [Google Scholar] [Crossref]

15. C. Chen, R. Jafari, and N. Kehtarnavaz, “UTD-MHAD: A Multimodal Dataset for Human Action Recognition Utilizing Depth and Inertial Sensors,” in IEEE International Conference on Image Processing (ICIP), 2015, pp. 168–172. [Google Scholar] [Crossref]

• An Adaptive Joint Filtering Approach to Wireless Relay Network for Transmission Rate Maximization
• IoT-Integrated Mercury Substance Detection System for Cosmetic Product Safety
• Design and Implementation of Solar PV-Based Railway Microgrid for Linke Hofmann Busch Coaches
• Cost Control Techniques on Civil Engineering Projects in Oyo State, Nigeria
• Strength and Predictive Modeling of Corn Cob Ash Blended Concrete Using Multi-Output Artificial Neural Network Approach