Abstract
Summary
Video-estimated peak jump power (vJP) using deep learning showed strong agreement with ground truth jump power (gJP). vJP was associated with sarcopenia, age, and muscle parameters in adults, with providing a proof-of-concept that markerless monitoring of peak jump power could be feasible in daily life space.
Objectives
Low peak countermovement jump power measured by ground force plate (GFP) is associated with sarcopenia, impaired physical function, and elevated risk of fracture in older adults. GFP is available at research setting yet, limiting its clinical applicability. Video-based estimation of peak jump power could enhance clinical applicability of jump power measurement over research setting.
Methods
Data were collected prospectively in osteoporosis clinic of Severance Hospital, Korea, between March and August 2022. Individuals performed three jump attempts on GFP (ground truth, gJP) under video recording, along with measurement of handgrip strength (HGS), 5-time chair rise (CRT) test, and appendicular lean mass (ALM). Open source deep learning pose estimation and machine learning algorithms were used to estimate video-estimated peak jump power (vJP) in 80% train set. Sarcopenia was defined by Korean Working Group for Sarcopenia 2023 definition.
Results
A total of 658 jump motion data from 220 patients (mean age 62 years; 77% women; sarcopenia 19%) were analyzed. In test set (20% hold-out set), average difference between predicted and actual jump power was 0.27 W/kg (95% limit of agreement − 5.01 to + 5.54 W/kg; correlation coefficient 0.93). vJP detected gJP-defined low jump power with 81.8% sensitivity and 91.3% specificity. vJP showed a steep decline across age like gJP, with modest to strong correlation with HGS and CRT. Eight landmarks (both shoulders, hip, knee joints, and ears) were the most contributing features to vJP estimation. vJP was associated with the presence of sarcopenia (unadjusted and adjusted, − 3.95 and − 2.30 W/kg), HGS (− 3.69 and − 1.96 W/kg per 1 SD decrement), and CRT performance (− 2.79 and − 1.87 W/kg per 1 SD decrement in log-CRT) similar to that of gJP.
Conclusion
vJP was associated with sarcopenia, age, and muscle parameters in adults, with good agreement with ground truth jump power.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data Availability
Data sharing requests will be considered for the research purpose upon a written request to the corresponding author. If agreed, deidentified participant data and/or deep learning model weights will be made available, subject to a data sharing agreement. The scripts for the test are available at the Github page (https://github.com/nkhong84/vJP_project).
References
Cawthon PM et al (2020) Putative cut-points in sarcopenia components and incident adverse health outcomes: an SDOC analysis. J Am Geriatr Soc 68:1429–1437. https://doi.org/10.1111/jgs.16517
Kirk B. et al. (2024) The conceptual definition of sarcopenia: Delphi consensus from the Global Leadership Initiative in Sarcopenia (GLIS). Age and Ageing 53:afae052 https://doi.org/10.1093/ageing/afae052
Buehring B et al (2015) Reproducibility of jumping mechanography and traditional measures of physical and muscle function in older adults. Osteoporos Int 26:819–825. https://doi.org/10.1007/s00198-014-2983-z
Hong N, Kim CO, Youm Y, Kim HC, Rhee Y (2018) Low peak jump power is associated with elevated odds of dysmobility syndrome in community-dwelling elderly individuals: the Korean Urban Rural Elderly (KURE) study. Osteoporos Int 29:1427–1436. https://doi.org/10.1007/s00198-018-4466-0
Hong N et al (2021) Defining an international cut-off of two-legged countermovement jump power for sarcopenia and dysmobility syndrome. Osteoporos Int 32:483–493. https://doi.org/10.1007/s00198-020-05591-x
Hong N, Burm SW, Kim HC, Kim CO, Rhee Y (2023) Jump power predicts fracture risk in older adults independent of sarcopenia and FRAX. J Bone Miner Res 38:958–967. https://doi.org/10.1002/jbmr.4825
Kidziński Ł et al (2020) Deep neural networks enable quantitative movement analysis using single-camera videos. Nat Commun 11:4054. https://doi.org/10.1038/s41467-020-17807-z
Rittweger J, Schiessl H, Felsenberg D, Runge M (2004) Reproducibility of the jumping mechanography as a test of mechanical power output in physically competent adult and elderly subjects. J Am Geriatr Soc 52:128–131. https://doi.org/10.1111/j.1532-5415.2004.52022.x
Veilleux LN, Rauch F (2010) Reproducibility of jumping mechanography in healthy children and adults. J Musculoskelet Neuronal Interact 10:256–266
Baek JY et al (2023) Korean Working Group on Sarcopenia guideline: expert consensus on sarcopenia screening and diagnosis by the Korean Society of Sarcopenia, the Korean Society for Bone and Mineral Research, and the Korean Geriatrics Society. Ann Geriatr Med Res 27:9–21. https://doi.org/10.4235/agmr.23.0009
Jung H-W, Baek JY (2023) New clinical practice guidelines for sarcopenia screening and diagnosis in Korean older adults: a step forward. Ann Geriatr Med Res 27:1–2. https://doi.org/10.4235/agmr.23.0033
Kalman RE (1960) A new approach to linear filtering and prediction problems.
Ke G et al. (2017) Lightgbm: a highly efficient gradient boosting decision tree. Adv neural inf process syst 30
Royston P, Altman DG (1994) Regression using fractional polynomials of continuous covariates: parsimonious parametric modelling. J R Stat Soc. Ser C (Appl Stat) 43:429–467. https://doi.org/10.2307/2986270
Lundberg SM, Lee S-I (2017) in Advances in neural information processing systems vol. 30 (eds I. Guyon et al.) (Curran Associates, Inc.,)
Boswell MA et al. (2023) Smartphone videos of the sit-to-stand test predict osteoarthritis and health outcomes in a nationwide study. npj Digit Med 6: 32. https://doi.org/10.1038/s41746-023-00775-1
Beddiar DR, Oussalah M, Nini B (2022) Fall detection using body geometry and human pose estimation in video sequences. J Vis Commun Image Represent 82:103407. https://doi.org/10.1016/j.jvcir.2021.103407
Chen Z, Wang Y, Yang W. in Big data. (eds Xiangke Liao et al.) 283–296 (Springer Singapore)
Huang Z, Liu Y, Fang Y, Horn BKP in 2018 4th International Conference on Universal Village (UV). 1–4
Wu L et al. (2023) Video-based fall detection using human pose and constrained generative adversarial network. IEEE Trans Circ Syst for Video Technol 1–1. https://doi.org/10.1109/TCSVT.2023.3303258
Hu R et al. (2024) Effective evaluation of HGcnMLP method for markerless 3D pose estimation of musculoskeletal diseases patients based on smartphone monocular video. Front Bioeng Biotechnol 11. https://doi.org/10.3389/fbioe.2023.1335251
Kim S et al (2023) Assessing physical abilities of sarcopenia patients using gait analysis and smart insole for development of digital biomarker. Sci Rep 13:10602. https://doi.org/10.1038/s41598-023-37794-7
Stenum J, Rossi C, Roemmich RT (2021) Two-dimensional video-based analysis of human gait using pose estimation. PLoS Comput Biol 17:e1008935. https://doi.org/10.1371/journal.pcbi.1008935
Dulhazi F, Agod S, Szarka S, Tuza K, Kopper B (2024) AI based motion analysis software for sport and physical therapy assessment. Revista Brasileira de Medicina do Esporte 30. https://doi.org/10.1590/1517-8692202430012022_0020i
Zheng C et al. (2023) Deep learning-based human pose estimation: a survey. ACM Comput Surv 56:Article 11. https://doi.org/10.1145/3603618
Cao Z, Hidalgo G, Simon T, Wei SE, Sheikh Y (2021) OpenPose: realtime multi-person 2D pose estimation using part affinity fields. IEEE Trans Pattern Anal Mach Intell 43:172–186. https://doi.org/10.1109/TPAMI.2019.2929257
Fang H-S et al (2023) AlphaPose: whole-body regional multi-person pose estimation and tracking in real-time. IEEE Trans Pattern Anal Mach Intell 45:7157–7173. https://doi.org/10.1109/tpami.2022.3222784
Wang Y et al (2023) The willingness to continue using wearable devices among the elderly: SEM and FsQCA analysis. BMC Med Inform Decis Mak 23:218. https://doi.org/10.1186/s12911-023-02336-8
Welch, G. & Bishop, G. in International conference on computer graphics and interactive techniques.
Stanton R, Kean CO, Scanlan AT (2015) My Jump for vertical jump assessment. Br J Sports Med 49:1157–1158. https://doi.org/10.1136/bjsports-2015-094831
Gençoğlu C et al (2023) Validity and reliability of “My Jump app” to assess vertical jump performance: a meta-analytic review. Sci Rep 13:20137. https://doi.org/10.1038/s41598-023-46935-x
Funding
This research was supported by Young Medical Scientist Research Grant through the Seokchunnanum Foundation (SCY2104P) and by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (2022R1 C1 C1006807). The funding sources had no involvement in study design, data collection, analysis, interpretation of data, writing of the report, and decision to submit the paper.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
None.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
ESM1
(PDF 657 KB)
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Cho, S.W., Cho, S.J., Park, Ey. et al. Video-estimated peak jump power using deep learning is associated with sarcopenia and low physical performance in adults. Osteoporos Int 36, 1193–1201 (2025). https://doi.org/10.1007/s00198-025-07515-z
Received:
Accepted:
Published:
Issue date:
DOI: https://doi.org/10.1007/s00198-025-07515-z