TheraCoordination: um sistema multimídia de realidade virtual com análise automática de movimento para reabilitação motora

• Eucila Gisele Silva de Freitas UFPA
• Clara de Aquino Rodrigues UFPA
• Tobias Moraes de Souza UFPA
• Ramon Costa de Lima UFPA
• Marcos Cesar da Rocha Seruffo UFPA
• Frederico Guilherme Santana da Silva Filho UFPA

DOI: https://doi.org/10.5753/webmedia.2026.30952

Resumo

This paper presents TheraCoordination, a virtual reality (VR) proof of concept that integrates the Meta Quest 3 with hand tracking, on-device motion classification (Random Forest quantized to 8-bit via Unity Sentis), and a web dashboard for monitoring. The model achieved 80.2% accuracy for gesture recognition and 87.6% for quality classification, with a size of 1.2 MB and inference time of 3 ms. The system computes kinematic metrics (accuracy, speed, ROM) in real time, maintaining 71.8 fps and latency < 20 ms. The results demonstrate the technical feasibility of a quantized ML pipeline on a standalone headset, providing objective data for clinical monitoring. Patient validation is in the planning phase.

Palavras-chave: virtual reality, motor rehabilitation, machine learning, hand tracking, movement classification

Referências

Ahmed Abdlkarim et al. 2024. Markerless hand tracking for clinical assessment: Validation and application in rehabilitation. IEEE Transactions on Biomedical Engineering 71, 2 (2024), 456–468. DOI: 10.1109/TBME.2024.3156789

Ahmed Abdlkarim et al. 2024. A methodological framework to assess the accuracy of virtual reality hand-tracking systems: A case study with the Meta Quest 2. Behavior Research Methods 56, 2 (2024), 1052–1063. DOI: 10.3758/s13428-022-02051-8

Sivakumar Balasubramanian, Alejandro Melendez-Calderon, Agnes Roby-Brami, and Etienne Burdet. 2015. On the analysis of movement smoothness. Journal of NeuroEngineering and Rehabilitation 12 (2015), 112. DOI: 10.1186/s12984-015-0090-9

Margherita Ceradini, Emilia Losanno, Silvestro Micera, and Andrea Bandini. 2024. Immersive VR for upper-extremity rehabilitation in patients with neurological disorders: A scoping review. Journal of NeuroEngineering and Rehabilitation 21, 1 (2024), 76. DOI: 10.1186/s12984-024-01367-0

Ke Cheng, Yifan Zhang, Xiangyu He, Weihan Chen, Jian Cheng, and Hanqing Lu. 2021. Spatial-Temporal Transformer for Skeleton-Based Action Recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35. 1187–1195.

Kate E. Laver, Belinda Lange, Stacey George, Judith E. Deutsch, Gustavo Saposnik, and Maria Crotty. 2017. Virtual reality for stroke rehabilitation. Cochrane Database of Systematic Reviews 11 (2017), CD008349. DOI: 10.1002/14651858.CD008349.pub4

Yalin Liao, Aleksandar Vakanski, and Min Xian. 2020. A Deep Learning Framework for Assessing Physical Rehabilitation Exercises. IEEE Transactions on Neural Systems and Rehabilitation Engineering 28, 2 (2020), 468–477. DOI: 10.1109/TNSRE.2020.2966249

Virgil Mathiowetz, Gloria Volland, Nancy Kashman, and Karen Weber. 1985. Adult Norms for the Box and Block Test of Manual Dexterity. The American Journal of Occupational Therapy 39, 6 (1985), 386–391. DOI: 10.5014/ajot.39.6.386

Alina Miron, Noureddin Sadawi,Waidah Ismail, Hafez Hussain, and Crina Grosan. 2021. IntelliRehabDS (IRDS)—A Dataset of Physical Rehabilitation Movements. Data 6, 5 (2021), 46. DOI: 10.3390/data6050046

Francisco-Javier Peláez-Vélez, Martina Eckert, Mariano Gacto-Sánchez, and Angel Martinez-Carrasco. 2024. Use of virtual reality and videogames in the physiotherapy treatment of stroke patients: A pilot randomized controlled trial. JMIR Serious Games (2024).

Gustavo Saposnik, Robert Teasell, Muhammad Mamdani, Judith Hall, William McIlroy, Donna Cheung, Kevin E. Thorpe, Leonardo G. Cohen, and Mark Bayley. 2010. Effectiveness of virtual reality using Wii gaming technology in stroke rehabilitation: A pilot randomized clinical trial and proof of principle. Stroke 41, 7 (2010), 1477–1484. DOI: 10.1161/STROKEAHA.110.584979

Mohsen Soleimani, Marjan Ghazisaeedi, and Soroush Heydari. 2024. The efficacy of virtual reality for upper limb rehabilitation in stroke patients: a systematic review and meta-analysis. BMC Medical Informatics and Decision Making 24 (2024), 135. DOI: 10.1186/s12911-024-02534-y

Aleksandar Vakanski, Hyung-pil Jun, David Paul, and Russell Baker. 2018. A Data Set of Human Body Movements for Physical Rehabilitation Exercises. Data 3, 1 (2018), 2. DOI: 10.3390/data3010002

Sijie Yan, Yuanjun Xiong, and Dahua Lin. 2018. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 32. DOI: 10.1609/aaai.v32i1.12328