A TCGAN-Based Real-Time Personalized Motion Guidance System to Reduce Compensatory Movements in Post-Stroke Rehabilitation

Authors

  • Fouzi Lezzar LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria https://orcid.org/0000-0002-9520-4924
  • Djamel Benmerzoug LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria https://orcid.org/0000-0002-6682-2862
  • Souheila Boudouda LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria https://orcid.org/0009-0000-6993-2533
  • Mohamed Lamine Berkane LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria https://orcid.org/0000-0001-8781-0257
  • Seif Eddine Mili Ecole Normale Supérieure, Constantine, Engineering Laboratory for Complex Systems (LISCO), Annaba University, Algeria https://orcid.org/0000-0002-8424-3354

DOI:

https://doi.org/10.5755/j01.itc.54.3.40634

Keywords:

Home-based rehabilitation, compensation assessment, real-time exercise guidance, TCGAN, post-stroke recovery

Abstract

Stroke rehabilitation is essential for motor function recovery, yet traditional methods require therapist supervision, which can be costly and inaccessible. Home-based rehabilitation offers an alternative, but without real-time guidance, patients may develop compensatory movements, hindering progress. Existing approaches provide feedback only after exercises are completed, limiting their effectiveness. To address this, we propose a Temporal Conditional Generative Adversarial Network (TCGAN)-based motion generation system that provides real-time skeletal guidance tailored to each patient’s body structure and positioning. By detecting key anatomical landmarks and generating adaptive motion sequences, the system ensures precise movement execution, reducing errors and improving rehabilitation outcomes. Both qualitative and quantitative evaluations confirm the effectiveness of the generated exercises, benefiting from the proposed architecture, improved loss function, optimized training process, and TCGAN hyperparameter tuning. Experimental results show a high degree of similarity between generated and real movements, with a Fréchet Inception Distance (FID) score of 0.87, demonstrating the system’s realism and reliability. This approach enhances patient autonomy and recovery efficiency, offering a more interactive and adaptive rehabilitation experience.    

Author Biographies

  • Fouzi Lezzar, LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria

    Fouzi Lezzar is currently an assistant professor in the department of TLSI, Faculty of New Technologies of Information and Communication, University Constantine 2, Algeria. He holds a PhD in Computer science from Batna 2 University (Batna - Algeria). His research interest includes Artificial intelligence, Internet of Things, smart cities and ehealth.

  • Djamel Benmerzoug, LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria

    Received the Ph.D. degree in computer science from Pierre and Marie Curie University, Paris, France. He is currently a Full Professor with the Department of TLSI, Faculty of New Technologies of Information and Communication, University Constantine 2, Algeria. He has published many papers in many international conferences and journals. He supervises many master’s and Ph.D. students. His current research interests include the Internet of Things, cloud computing, advanced enterprises systems, multiagent systems, service oriented computing, and business processes modeling and verification.

  • Souheila Boudouda, LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria

    Souheila Boudouda is an Assistant Professor at the Software Technologies and Information Systems Department of Abdelhamid Mehri Constantine-2 University in Algeria. She is a Research Associate in the Research group "Information Systems and Knowledge Bases", LIRE laboratory, Abdelhamid Mehri Constantine 2 University, Algeria (https://lire-lab.com/). Her current research interests include Supply Chain Management, Cooperative Systems, E-business applications, Artificial Intelligent and IoT in Smart City.

  • Mohamed Lamine Berkane, LIRE Laboratory, Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri Constantine 2, Constantine 25016, Algeria

    Mohamed Lamine Berkane is a member of LIRE Laboratory at the Abdelhamid Mehri – Constantine 2 (Software Technologies and Information Systems Department) University and received her PhD in Computer Science in 2015. His research interests include self-adaptive systems, cloud computing and green computing.

  • Seif Eddine Mili, Ecole Normale Supérieure, Constantine, Engineering Laboratory for Complex Systems (LISCO), Annaba University, Algeria

    Seif Eddine Mili is an assistant professor in the Department of Computer Science, Ecole Normale supérieure Constantine (Department of Mathematics), Algeria. He holds a PhD in Computer science from Badji Mokhtar Annaba University, Algeria. His research interest includes Artificial intelligence, Web of Things, IoT, Multi Agent Systems; Bio inspired System and Model Driven Architecture.

Downloads

Published

2025-10-08

Issue

Section

Articles