Artificial intelligence in postural management: a critical review of detection, correction, and clinical applicability
Искусственный интеллект в управлении осанкой: критический обзор выявления, коррекции и клинической применимости
2026-01-21
SCID: 54.1/bkwqw6ws
Discuss with AI
artificial intelligencecomputer visionhuman pose estimationpostural managementtele-rehabilitation
Figures from the paper
Abstract (AI)
BACKGROUND: Poor posture and related musculoskeletal conditions represent a growing global health concern. Conventional postural assessment methods are often subjective, intermittent, and insufficient for accurate, continuous monitoring. Advances in artificial intelligence (AI), particularly in computer vision and human pose estimation (HPE), have introduced new possibilities for objective and real-time postural analysis. MAIN BODY: This critical review synthesizes and evaluates current developments in AI technologies for postural management. The review draws on recent literature from computer science, bioengineering, and clinical research, focusing on studies from the past decade that explore the use of AI and HPE in the detection, monitoring, and correction of human posture. AI-based HPE models demonstrate high precision in identifying anatomical landmarks and quantifying postural parameters, offering a robust alternative to traditional assessment methods. Applications are expanding beyond laboratory environments to practical contexts such as ergonomic risk evaluation and sports performance analysis. In addition, AI-driven systems that deliver real-time feedback and support tele-rehabilitation are enhancing user engagement and enabling personalized interventions. Despite these advancements, the field faces several challenges. Evidence from large-scale clinical trials remains limited, and the generalizability of existing models across diverse populations and real-world conditions is uncertain. Concerns related to usability, data privacy, and integration within healthcare systems also pose significant barriers to clinical translation. CONCLUSION: AI holds considerable potential to transform postural management through continuous, objective, and accessible assessment and intervention. To fully realize this potential, future work must extend beyond technical innovation to include rigorous clinical validation, user-centered design, and the establishment of ethical and regulatory frameworks that ensure safe, effective, and equitable implementation.
Key Findings
1
AI postural analysis is expanding from laboratory settings to ergonomic risk evaluation and sports performance applications.
2
AI-based human pose estimation can identify anatomical landmarks and quantify postural parameters with high precision, providing a more objective alternative to conventional assessment.
3
Large-scale clinical evidence remains limited, and model generalizability across diverse populations and real-world conditions is uncertain.
4
Real-time AI feedback and tele-rehabilitation systems can improve user engagement and support personalized postural interventions.
5
Usability, data privacy, healthcare-system integration, and the need for rigorous clinical validation and user-centered design remain major barriers to clinical translation.
Research Object
Human posture and postural management in clinical, ergonomic, sports, and rehabilitation contexts
Research Subject
AI- and human pose estimation–based detection, monitoring, correction, and clinical applicability of postural abnormalities
Publication Details
Publication Date
2026-01-21
Journal
Publisher
ISSN
Cited by
9
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest