Authors :
Raghav Gupta; Sanjay Srivastava; Priyanshu Srivastava; Rohit Maurya; Shubham Singh Rao
Volume/Issue :
Volume 11 - 2026, Issue 4 - April
Google Scholar :
https://tinyurl.com/276bks63
Scribd :
https://tinyurl.com/4bmda7kz
DOI :
https://doi.org/10.38124/ijisrt/26apr1865
Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.
Abstract :
Patients are expected to perform the necessary rehabilitation exercises with precision and reliability following
any injury, surgery, or case of a neurological condition if they are to produce the best possible results in their physiotherapy.
Traditional physiotherapy faces three challenges: lack of professional supervision at every point in time, abuse of the
exercises done in the home setting, and lack of consistency in checking progress. To solve these problems, the
recommendation presented in the article was that there should be a "Kinetic Master," which is an artificial intelligence
system for the monitoring of patient rehabilitation. It provides only automatic and immediate instructions for the training
of the patients. In the proposed system for the patients, the patients are given immediate feedback for the exercises carried out
by them to ensure that the patients are in the right position throughout the exercises. The prediction function associated with
the use of predictive analytics also createsspace for another use, where the use of the same technique could be applied for the
improvement of patient monitoring for the planning of therapies carried out by the patients. For instance, the proposed
system for the patients could be more accurate and efficient for the monitoring of the patients to ensure the promotion of the
care associated with the scalable system for the promotion of healthcare.
Keywords :
Rehabilitation Monitoring, Computer Vision, Machine Learning, Remote Patient Monitoring.
References :
- S. Z. Li, Q. Liu, H. Song, and Y. Zhao, "Rehabilitation Exercises Monitoring Using Pose Estimation and Deep Learning Models," IEEE Access, vol. 10, pp. 116450–116462, 2022.
- H. Wang, R. Kumar, and F. Chen, "AI-Assisted Physiotherapy: A Review of Current Trends and Challenges," SN Appl. Sci., vol. 5, no. 4, pp. 234–247, 2023.
- P. Gupta, A. Sharma, and M. Singh, "Real-Time Feedback Systems for Home-Based Rehabilitation Using AI and IoT," ACM Trans. Comput. Healthcare, vol. 4, no. 2, pp. 1–19, 2023.
- Z. Cao et al., "OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields," IEEE Trans. Pattern Anal. Mach. Intell., vol. 43, no. 1, pp. 172–186, 2021.
- Google, "MediaPipe Pose," 2021. [Online]. Available: https://mediapipe.dev
- A. Tharatipyakul et al., "Deep Learning-Based Human Body Pose Estimation in Healthcare: A Review," Biomed. Signal Process. Control, vol. 82, 2024.
- K. Wang et al., "TFC-GCN: Temporal Feature Cross-Extraction GCN for Skeleton-Based Action Recognition," Appl. Intell., vol. 53, pp. 14521– 14535, 2023.
- Y. Kim, J. Park, and H. Lee, "Reinforcement Learning for Personalized Rehabilitation," IEEE Trans. Neural Syst. Rehabil. Eng., vol. 32, no. 5, pp. 1021–1032, 2024.
- H. Wang, R. Kumar, and F. Chen, "AI-Assisted Physiotherapy: A Review of Current Trends and Challenges," SN Appl. Sci., 2023.
- M. Haghi, K. Thurow, and R. Stoll, "Wearable Devices in Medical Internet of Things," Healthcare Informatics Research, vol. 23, no. 1, pp. 4– 15, 2017.
- WHO, "Digital Health Interventions for Health System Strengthening," 2019.
- Google, "MediaPipe Pose," 2021.
- P. Gupta et al., "Real-Time Feedback Systems for Home-Based Rehabilitation," 2023.
- P. Gupta et al., "AI-based Engagement in Rehabilitation Systems," 2023.
- S. Z. Li et al., "Rehabilitation Monitoring Using AI Feedback Systems," 2022.
- A. Tharatipyakul et al., "Deep Learning Pose Estimation in Healthcare," 2024.
- S. Sicari et al., "Security, Privacy and Trust in IoT," Computer Networks, 2015.
- X. Dai et al., "Full-Body Pose Reconstruction and Correction in Virtual Rehabilitation," Front. Neurosci., 2024.
- R. Hang et al., "Spatial-Temporal GCN for Action Recognition," ACCV, 2022.
- H. S. Fang et al., "Whole-Body Pose Estimation," arXiv, 2022.
- J. Donahue et al., "Long-Term Recurrent Convolutional Networks," IEEE TPAMI, 2015.
- X. Shi et al., "Skeleton-Based Action Recognition Using LSTM," IEEE TIP, 2017.
- T. Tomy et al., "Yoga Pose Estimation Using Deep Learning," 2024.
Patients are expected to perform the necessary rehabilitation exercises with precision and reliability following
any injury, surgery, or case of a neurological condition if they are to produce the best possible results in their physiotherapy.
Traditional physiotherapy faces three challenges: lack of professional supervision at every point in time, abuse of the
exercises done in the home setting, and lack of consistency in checking progress. To solve these problems, the
recommendation presented in the article was that there should be a "Kinetic Master," which is an artificial intelligence
system for the monitoring of patient rehabilitation. It provides only automatic and immediate instructions for the training
of the patients. In the proposed system for the patients, the patients are given immediate feedback for the exercises carried out
by them to ensure that the patients are in the right position throughout the exercises. The prediction function associated with
the use of predictive analytics also createsspace for another use, where the use of the same technique could be applied for the
improvement of patient monitoring for the planning of therapies carried out by the patients. For instance, the proposed
system for the patients could be more accurate and efficient for the monitoring of the patients to ensure the promotion of the
care associated with the scalable system for the promotion of healthcare.
Keywords :
Rehabilitation Monitoring, Computer Vision, Machine Learning, Remote Patient Monitoring.
