LRLiva Redliha

Liva Redliha

*This is an algorithmically generated author record based on publicly available information.

Liva Redliha has a total of 2 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Leveraging AI and Machine Vision in Telephysiotherapy for Promoting Physical Activity and Reducing Sedentary Behavior: Scoping Review

    by Liva Redliha, Guna Semjonova, Liva Araka

    Background: Globally, sedentary behavior is a leading risk factor for noncommunicable diseases and premature death. Although telerehabilitation has emerged as a promising method to increase access to physical therapy, the potential of artificial intelligence (AI) and machine vision technologies to objectively monitor and promote physical activity (PA) in home settings remains underexplored. Objective: This scoping review aimed to map and systematize current evidence on the use of AI and machine vision in telephysiotherapy to promote PA and reduce sedentary behavior among healthy adults. Methods: The review followed the Joanna Briggs Institute scoping review methodology and was reported according to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. The PCC (population, concepts, and context) framework defined participants as healthy adults, the concept as AI or machine learning (ML) and machine vision technologies, and the context as remote, home-based telephysiotherapy for objective PA monitoring and/or outcome measurement in interventions aimed at promoting daily activity. Searches were conducted in PubMed, ProQuest, Science Direct, and Wiley Online Library for peer-reviewed primary studies published between 2016 and 2024. Inclusion and exclusion criteria were applied systematically using the Rayyan platform. Results: Of 7253 initial records, 4 studies met the inclusion criteria. Identified digital solutions comprised ML algorithms (eg, extreme gradient boosting and hybrid classifiers), deep learning models such as graph convolutional networks, video-based center-of-pressure assessments using RGB cameras, and ontology-driven semantic data structuring for personalization. The most frequently studied technologies combined wearable sensors with ML-based behavior change frameworks to assess usability and prototype feasibility. However, none of the studies demonstrated statistically significant improvements in reducing sedentary time or increasing PA levels among participants. The studies predominantly used exploratory or pilot designs, with small sample sizes (range: 10‐32 participants) and short follow-up periods (≤12 wk), limiting the strength of the conclusions. Conclusions: Current evidence on the application of AI and machine vision in telephysiotherapy remains scarce and largely exploratory. Although the integration of wearable technology and ML algorithms shows technical feasibility and personalization potential, there is insufficient evidence to confirm improvements in PA outcomes. Future research should prioritize longitudinal clinical trials, ensure the inclusion of diverse populations, and evaluate behavioral outcomes to determine the real-world impact of these technologies in telerehabilitation.

  • Open Access

    Leveraging AI and Machine Vision in Telephysiotherapy for Promoting Physical Activity and Reducing Sedentary Behavior: A Scoping Review (Preprint)

    by Liva Redliha, Guna Semjonova, Liva Araka

    BACKGROUND Globally, sedentary behavior is a leading risk factor for non-communicable diseases and premature death. Although telerehabilitation has emerged as a promising method to increase access to physical therapy, the potential of artificial intelligence (AI) and machine vision technologies to objectively monitor and promote physical activity (PA) in home settings remains underexplored. OBJECTIVE This scoping review aimed to map and systematize current evidence on the use of AI and machine vision in telephysiotherapy to promote physical activity and reduce sedentary behavior among healthy adults. METHODS The review followed the Joanna Briggs Institute scoping review methodology and was reported according to the PRISMA-ScR checklist. The PCC framework defined participants as healthy adults, the concept as AI/machine learning (ML) and machine vision technologies, and the context as remote, home-based telephysiotherapy for objective PA monitoring and/or outcome measurement in interventions aimed at promoting daily activity. Searches were conducted in PubMed, ProQuest, Science Direct, and Wiley Online Library for peer-reviewed primary studies published between 2016 and 2024. Inclusion and exclusion criteria were applied systematically using the Rayyan platform. RESULTS Of 7253 initial records, 4 studies met the inclusion criteria. Identified digital solutions included: Machine learning algorithms (e.g., XGBoost, hybrid classifiers), Deep learning models such as Graph Convolutional Networks (GCN), Video-based Center of Pressure (CoP) assessments using RGB cameras, Ontology-driven semantic data structuring for personalization. The most frequently studied technologies combined wearable sensors with ML-based behavioral change frameworks to assess usability and prototype feasibility. However, none of the studies demonstrated statistically significant improvements in reducing sedentary time or increasing PA levels among participants. The studies predominantly employed exploratory or pilot designs, with small sample sizes (range: 10–32 participants), and short follow-up periods (≤ 12 weeks), limiting the strength of conclusions. CONCLUSIONS Current evidence on the application of AI and machine vision in telephysiotherapy remains scarce and largely exploratory. Although the integration of wearable technology and ML algorithms shows technical feasibility and personalization potential, there is insufficient evidence to confirm improvements in physical activity outcomes. Future research should prioritize longitudinal clinical trials, include diverse populations, and evaluate behavioral outcomes to determine the real-world impact of these technologies in telerehabilitation. CLINICALTRIAL Semjonova, G., & Rēdliha, L. (2025, July 5). The Use of Machine Vision and Artificial Intelligence in Tele-Physiotherapy to Reduce Sedentary Lifestyles and Promote Physical Activity: A Scoping Review. https://doi.org/10.17605/OSF.IO/SFC97