EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Recognition of Sensory-based Motion Information using Virtual Reality based Rehabilitation Exercises for COVID19 Recovered Patients

Mirza Naveed Shahzad, Majid Harouni

International Journal of Theoretical & Applied Computational Intelligence · 2025

DOI: 10.65278/ijtaci.2025.19 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Theoretical & Applied Computational Intelligence, 2025. Verfasst von Mirza Naveed Shahzad und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Exercise therapy is regarded as one of the major treatments for the rehabilitation of patients affected by injury, mental disorder, and even COVID-19. Virtual reality in healthcare rehabilitation demands accurate classification of gestures or movements. This study aims to recognize the prescribed rehabilitation exercises of computer-assisted sensory systems using machine learning (ML) and deep learning (DL) approaches for the automated assessment of the quality of physical rehabilitation exercises in COVID-19 recovered patients. The Kinect and Vicon sensory systems are used in this study to capture ten movements of a rehabilitation program. Patients perform these ten movements in their physical therapy and rehabilitation program. The ML models K-Nearest Neighbour (KNN) and Decision Tree (DT), and the DL-based Multi-Layer Perceptron (MLP), are used to classify the Kinect and Vicon data. The binary classification performance is measured with standard methods and metrics. This study utilizes ten movements of ten subjects from two sensory systems, and the mean accuracy over the training and testing parts is 88.3%, 89.4%, and 78.95% for KNN, DT, and MLP, respectively. The overall mean accuracy of DT is higher than that of the other two models. The DT method shows better performance than the other methods, with the best-learned features for any rehabilitation exercise. The DT reveals that the most important rehabilitation exercises for recognizing the Kinect and Vicon sensory system information are deep squats and sit-to-stands. This study achieves an overall accuracy of 89.4% in recognizing prescribed rehabilitation exercises of computer-assisted sensory systems. Therefore, the findings are helpful in clinical gait analysis, medical applications, and related computer-assisted sensory systems for the recognition process.

Zitieren

APA

Shahzad, M. N., & Harouni, M. (2025). Recognition of Sensory-based Motion Information using Virtual Reality based Rehabilitation Exercises for COVID19 Recovered Patients. International Journal of Theoretical & Applied Computational Intelligence. https://doi.org/10.65278/ijtaci.2025.19

Vancouver

Shahzad MN, Harouni M. Recognition of Sensory-based Motion Information using Virtual Reality based Rehabilitation Exercises for COVID19 Recovered Patients. International Journal of Theoretical & Applied Computational Intelligence. 2025. doi:10.65278/ijtaci.2025.19

BibTeX
@article{Shahzad2025,
  author = {Shahzad, Mirza Naveed and Harouni, Majid},
  title = {Recognition of Sensory-based Motion Information using Virtual Reality based Rehabilitation Exercises for COVID19 Recovered Patients},
  journal = {International Journal of Theoretical & Applied Computational Intelligence},
  year = {2025},
  doi = {10.65278/ijtaci.2025.19}
}
Kostenlos starten →