<![CDATA[Human pose estimation has become increasingly important in healthcare applications such as fall detection, gait analysis, and rehabilitation monitoring. However, existing systematic reviews remain fragmented and largely descriptive, with limited comparative benchmarking and insufficient attention to clinical validation. This study addresses this gap by providing a structured comparison of major pose estimation frameworks in healthcare contexts. A systematic literature review was conducted using the PICOC framework and PRISMA guidelines. Studies published between 2020 and 2025 were retrieved from Scopus, Web of Science, IEEE Xplore, and PubMed based on predefined inclusion and exclusion criteria. Following screening and quality assessment, 41 studies were included in the final analysis. The results indicate that framework performance varies according to application requirements. OpenPose offers high anatomical precision but requires substantial computational resources, whereas MoveNet and MediaPipe enable real-time performance with lower latency, making them suitable for mobile and telehealth settings. Nevertheless, the evidence remains heterogeneous, with challenges related to occlusion, lighting variability, lack of standardized datasets, and limited real-world clinical validation. This study contributes by providing a theoretical synthesis and practical guidance for selecting appropriate pose estimation frameworks in healthcare applications.]]>
Copyrights © 2026
