EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Application of machine learning for detecting high fall risk in middle-aged workers using video-based analysis of the first 3 steps

Naoki Sakane iD, Ken Yamauchi, Ippei Kutsuna, Akiko Suganuma, Masayuki Domichi, Kei Hirano, Kengo Wada, Masashi Ishimaru, Mitsuharu Hosokawa, Yosuke Izawa, Yoshihiro Matsumura, Junichi Hozumi

Journal of Occupational Health · 2025

DOI: 10.1093/joccuh/uiae075 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Occupational Health, 2025. Verfasst von Naoki Sakane und 11 weiteren Autor:innen. Beteiligte Einrichtung: Division of Preventive Medicine, Clinical Research Institute, National Hospital Organization Kyoto Medical Center, 1-1 Mukaihata-cho, Fukakusa, Fushi…. Hauptthemen laut MEDLINE: Accidental Falls, Machine Learning, Accidents, Occupational. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Panasonic Corporation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

Falls are among the most prevalent workplace accidents, necessitating thorough screening for susceptibility to falls and customization of individualized fall prevention programs. The aim of this study was to develop and validate a high fall risk prediction model using machine learning (ML) and video-based first 3 steps in middle-aged workers.

Methods

Participants to provide training data (n = 190, mean [SD] age = 54.5 [7.7] years, 48.9% male) and validation data (n = 28, age = 52.3 [6.0] years, 53.6% male) were enrolled in this study. Pose estimation was performed using a marker-free deep pose estimation method called MediaPipe Pose. The first 3 steps, including the movements of the arms, legs, trunk, and pelvis, were recorded using an RGB camera, and the gait features were identified. Using these gait features and fall histories, a stratified k-fold cross-validation method was used to ensure balanced training and test data, and the area under the curve (AUC) and 95% CI were calculated.

Results

Of 77 gait features in the first 3 steps, we found 3 gait features in men with an AUC of 0.909 (95% CI, 0.879-0.939) for fall risk, indicating an “excellent” (0.9-1.0) classification, whereas we determined 5 gait features in women with an AUC of 0.670 (95% CI, 0.621-0.719), indicating a “sufficient” (0.6-0.7) classification.

Conclusions

These findings suggest that fall risk prediction can be developed based on ML and the first 3 steps in men; however, the accuracy was only “sufficient” in women. Further development of the formula for women is required to improve its accuracy in the middle-aged working population.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsMachine LearningAccidents, OccupationalHumansMaleFemaleMiddle AgedVideo RecordingRisk AssessmentGaitRisk FactorsAdult

Stichwörter der Autor:innen

accidental fallsfallsgait analysismachine learningmiddle agerisk assessmentworkplace

Autor:innen und Institutionen

  • Division of Preventive Medicine, Clinical Research Institute, National Hospital Organization Kyoto Medical Center, 1-1 Mukaihata-cho, Fukakusa, Fushimi-ku, Kyoto 612-8555, Japan.
  • Institute of Physical Education, Keio University, 4-1-1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa 223-8521, Japan.
  • Department of Electric Works Company/Engineering Division, Panasonic Corporation,1006, Kadoma, Kadoma City, Osaka 571-8501, Japan.

ORCID-Profile: Naoki Sakane

Förderung

  • Panasonic Corporation

Zitieren

APA

Sakane, N., Yamauchi, K., Kutsuna, I., Suganuma, A., Domichi, M., Hirano, K., Wada, K., Ishimaru, M., Hosokawa, M., Izawa, Y., Matsumura, Y., & Hozumi, J. (2025). Application of machine learning for detecting high fall risk in middle-aged workers using video-based analysis of the first 3 steps. Journal of Occupational Health. https://doi.org/10.1093/joccuh/uiae075

Vancouver

Sakane N, Yamauchi K, Kutsuna I, Suganuma A, Domichi M, Hirano K, et al. Application of machine learning for detecting high fall risk in middle-aged workers using video-based analysis of the first 3 steps. Journal of Occupational Health. 2025. doi:10.1093/joccuh/uiae075

BibTeX
@article{Sakane2025,
  author = {Sakane, Naoki and Yamauchi, Ken and Kutsuna, Ippei and Suganuma, Akiko and Domichi, Masayuki and Hirano, Kei and Wada, Kengo and Ishimaru, Masashi and Hosokawa, Mitsuharu and Izawa, Yosuke and Matsumura, Yoshihiro and Hozumi, Junichi},
  title = {Application of machine learning for detecting high fall risk in middle-aged workers using video-based analysis of the first 3 steps},
  journal = {Journal of Occupational Health},
  year = {2025},
  doi = {10.1093/joccuh/uiae075}
}
Kostenlos starten →