Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model

<b>Background:</b> Occupational low back pain (LBP) is a pervasive health issue that significantly impacts productivity and contributes to work-related musculoskeletal disorders (WMSDs). Inadequate lifting postures are a primary, modifiable risk factor associated with LBP, making early d...

Full description

Saved in:
Bibliographic Details
Main Authors: Gee-Sern Jison Hsu, Jie Syuan Wu, Yin-Kai Dean Huang, Chun-Chieh Chiu, Jiunn-Horng Kang
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Life
Subjects:
Online Access:https://www.mdpi.com/2075-1729/15/3/358
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850090513053515776
author Gee-Sern Jison Hsu
Jie Syuan Wu
Yin-Kai Dean Huang
Chun-Chieh Chiu
Jiunn-Horng Kang
author_facet Gee-Sern Jison Hsu
Jie Syuan Wu
Yin-Kai Dean Huang
Chun-Chieh Chiu
Jiunn-Horng Kang
author_sort Gee-Sern Jison Hsu
collection DOAJ
description <b>Background:</b> Occupational low back pain (LBP) is a pervasive health issue that significantly impacts productivity and contributes to work-related musculoskeletal disorders (WMSDs). Inadequate lifting postures are a primary, modifiable risk factor associated with LBP, making early detection of unsafe practices crucial to mitigating occupational injuries. Our study aims to address these limitations by developing a markerless, smartphone-based camera system integrated with a deep learning model capable of accurately classifying lifting postures. <b>Material and Method:</b> We recruited 50 healthy adults who participated in lifting tasks using correct and incorrect postures to build a robust dataset. Participants lifted boxes of varying sizes and weights while their movements were recorded from multiple angles and heights to ensure comprehensive data capture. We used the OpenPose algorithm to detect and extract key body points to calculate relevant biomechanical features. These extracted features served as inputs to a bidirectional long short-term memory (LSTM) model, which classified lifting postures into correct and incorrect categories. <b>Results</b>: Our model demonstrated high classification accuracy across all datasets, with accuracy rates of 96.9% for Tr, 95.6% for the testing set, and 94.4% for training. We observed that environmental factors, such as camera angle and height, slightly influenced the model’s accuracy, particularly in scenarios where the subject’s posture partially occluded key body points. Nonetheless, these variations were minor, confirming the robustness of our system across different conditions. <b>Conclusions:</b> This study demonstrates the feasibility and effectiveness of a smartphone camera and AI-based system for lifting posture classification. The system’s high accuracy, low setup cost, and ease of deployment make it a promising tool for enhancing workplace ergonomics. This approach highlights the potential of artificial intelligence to improve occupational safety and underscores the relevance of affordable, scalable solutions in the pursuit of healthier workplaces.
format Article
id doaj-art-cff35b080e924a4385d68e5cafbb3a74
institution DOAJ
issn 2075-1729
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Life
spelling doaj-art-cff35b080e924a4385d68e5cafbb3a742025-08-20T02:42:32ZengMDPI AGLife2075-17292025-02-0115335810.3390/life15030358Automatic Detect Incorrect Lifting Posture with the Pose Estimation ModelGee-Sern Jison Hsu0Jie Syuan Wu1Yin-Kai Dean Huang2Chun-Chieh Chiu3Jiunn-Horng Kang4Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei 10607, TaiwanSchool of Medicine, College of Medicine, Taipei Medical University, Taipei 11031, TaiwanDepartment of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei 11031, TaiwanDepartment of Physical Medicine and Rehabilitation, Wan Fang Hospital, Taipei Medical University, Taipei 11696, TaiwanDepartment of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei 11031, Taiwan<b>Background:</b> Occupational low back pain (LBP) is a pervasive health issue that significantly impacts productivity and contributes to work-related musculoskeletal disorders (WMSDs). Inadequate lifting postures are a primary, modifiable risk factor associated with LBP, making early detection of unsafe practices crucial to mitigating occupational injuries. Our study aims to address these limitations by developing a markerless, smartphone-based camera system integrated with a deep learning model capable of accurately classifying lifting postures. <b>Material and Method:</b> We recruited 50 healthy adults who participated in lifting tasks using correct and incorrect postures to build a robust dataset. Participants lifted boxes of varying sizes and weights while their movements were recorded from multiple angles and heights to ensure comprehensive data capture. We used the OpenPose algorithm to detect and extract key body points to calculate relevant biomechanical features. These extracted features served as inputs to a bidirectional long short-term memory (LSTM) model, which classified lifting postures into correct and incorrect categories. <b>Results</b>: Our model demonstrated high classification accuracy across all datasets, with accuracy rates of 96.9% for Tr, 95.6% for the testing set, and 94.4% for training. We observed that environmental factors, such as camera angle and height, slightly influenced the model’s accuracy, particularly in scenarios where the subject’s posture partially occluded key body points. Nonetheless, these variations were minor, confirming the robustness of our system across different conditions. <b>Conclusions:</b> This study demonstrates the feasibility and effectiveness of a smartphone camera and AI-based system for lifting posture classification. The system’s high accuracy, low setup cost, and ease of deployment make it a promising tool for enhancing workplace ergonomics. This approach highlights the potential of artificial intelligence to improve occupational safety and underscores the relevance of affordable, scalable solutions in the pursuit of healthier workplaces.https://www.mdpi.com/2075-1729/15/3/358occupational back injurylifting posturecameramarkerless systempose estimationartificial intelligence
spellingShingle Gee-Sern Jison Hsu
Jie Syuan Wu
Yin-Kai Dean Huang
Chun-Chieh Chiu
Jiunn-Horng Kang
Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
Life
occupational back injury
lifting posture
camera
markerless system
pose estimation
artificial intelligence
title Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
title_full Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
title_fullStr Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
title_full_unstemmed Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
title_short Automatic Detect Incorrect Lifting Posture with the Pose Estimation Model
title_sort automatic detect incorrect lifting posture with the pose estimation model
topic occupational back injury
lifting posture
camera
markerless system
pose estimation
artificial intelligence
url https://www.mdpi.com/2075-1729/15/3/358
work_keys_str_mv AT geesernjisonhsu automaticdetectincorrectliftingposturewiththeposeestimationmodel
AT jiesyuanwu automaticdetectincorrectliftingposturewiththeposeestimationmodel
AT yinkaideanhuang automaticdetectincorrectliftingposturewiththeposeestimationmodel
AT chunchiehchiu automaticdetectincorrectliftingposturewiththeposeestimationmodel
AT jiunnhorngkang automaticdetectincorrectliftingposturewiththeposeestimationmodel