Home // GLOBAL HEALTH 2022, The Eleventh International Conference on Global Health Challenges // View article


Estimation of Lumbar Load from Web Image Using Convolutional Neural Network for Standing Forward Bending Stationary Posture

Authors:
Riku Nishimoto
Kyoko Shibata

Keywords: Deep learning, Lumbar load, Musculoskeletal model simulation.

Abstract:
To prevent lumbago, it is effective to understand one’s current posture situation and how to improve one’s posture. Therefore, this study proposes a method to constantly observe posture and evaluate the load on the body. The realization of this method visualizes the body loads in daily life and can be used as one of the means to maintain health. Hence, this study proposes to use a web camera for sensing and deep learning as a tool for estimating body load from web image. The body load is derived using a musculoskeletal model simulation based on the skeletal information extracted from the images, and a deep learning model is prepared in advance using this as the true value. In this paper, the first subject is a stationary standing forward bending posture. The accuracy of the created deep learning model is evaluated and the results indicate that the proposed estimation method can be useful in the case of a standing forward bending posture.

Pages: 13 to 16

Copyright: Copyright (c) IARIA, 2022

Publication date: November 13, 2022

Published in: conference

ISSN: 2308-4553

ISBN: 978-1-61208-995-9

Location: Valencia, Spain

Dates: from November 13, 2022 to November 17, 2022