Combining Postural Sway Parameters and Machine Learning to Assess Biomechanical Risk Associated with Load-Lifting Activities

Aug 2, 2026

Upload by Oliver

Long-term work-related musculoskeletal disorders are shaped by load-lifting duration, intensity, and repetition, yet conventional ergonomic assessment tools can be difficult to apply due to limited streamlined standardization. Wearable sensors paired with artificial intelligence offer an alternative for monitoring and reducing biomechanical risk. This study evaluates machine learning models trained on lumbar IMU-derived postural sway metrics to classify risk levels using the Revised NIOSH Lifting Equation. Results show that Gradient Boosting achieves the best performance, with high ROC-AUC under two validation strategies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

Introduction

Background and rationale

Aim and study approach

Methods

Participants and lifting scenarios

IMU-based postural sway parameter extraction

Machine learning models and validation

Results

Classification performance

Feature importance analysis

Discussion

Feasibility for occupational ergonomics

Conclusions

Article Combining Postural Sway Parameters and Machine Learning to Assess Biomechanical Risk Associated with Load-Lifting Activities Giuseppe Prisco 1, Maria Agnese Pirozzi 2, Antonella Santone 1, Mario Cesarelli 3, Fabrizio Esposito 2, Paolo Gargiulo 4, Francesco Amato 5 and Leandro Donisi 2, * Academic Editor: Dechang Chen Received: 4 December 2024 Revised: 27 December 2024 Accepted: 2 January 2025 Published: 4 January 2025 Citation: Prisco, G.; Pirozzi, M.A.; Santone, A.; Cesarelli, M.; Esposito, F.; Gargiulo, P.; Amato, F.; Donisi, L. Combining Postural Sway Parameters and Machine Learning to Assess Biomechanical Risk Associated with Load-Lifting Activities. Diagnostics 2025, 15, 105. [https://doi.org/](https://doi.org/)[ ](https://doi.org/)[10.3390/diagnostics15010105](10.3390/diagnostics15010105) [Copyright:](Copyright:) © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://creativecommons.org/](https://creativecommons.org/)[ ](https://creativecommons.org/)[licenses/by/4.0/](licenses/by/4.0/)) . 1 Department of Medicine and Health Sciences, University of Molise, 86100 Campobasso, Italy; [g.prisco2@studenti.unimol.it](g.prisco2@studenti.unimol.it) (G.P.); [antonella.santone@unimol.it](antonella.santone@unimol.it) (A.S.) 2 Department of Advanced Medical and Surgical Sciences, University of Campania Luigi Vanvitelli, 80138 Naples, Italy; [mariaagnese.pirozzi@unicampania.it](mariaagnese.pirozzi@unicampania.it) (M.A.P.); fabrizio.esposito@unicampania.it (F.E.) 3 Department of Engineering, University of Sannio, 82100 Benevento, Italy; mcesarelli@unisannio.it 4 Institute of Biomedical and Neural Engineering, Reykjavik University, 102 Reykjavik, Iceland; [paolo@ru.is](paolo@ru.is) 5 Department of Information Technology and Electrical Engineering, University of Naples Federico II, 80125 Naples, Italy; [framato@unina.it](framato@unina.it) * Correspondence: [leandro.donisi@unicampania.it](leandro.donisi@unicampania.it) Abstract: Background/Objectives: Long-term work-related musculoskeletal disorders are predominantly influenced by factors such as the duration, intensity, and repetitive nature of load lifting. Although traditional ergonomic assessment tools can be effective, they are often challenging and complex to apply due to the absence of a streamlined, standardized framework. Recently, integrating wearable sensors with artificial intelligence has emerged as a promising approach to effectively monitor and mitigate biomechanical risks. This study aimed to evaluate the potential of machine learning models, trained on postural sway metrics derived from an inertial measurement unit (IMU) placed at the lumbar region, to classify risk levels associated with load lifting based on the Revised NIOSH Lifting Equation. Methods: To compute postural sway parameters, the IMU captured acceleration data in both anteroposterior and mediolateral directions, aligning closely with the body’s center of mass. Eight participants undertook two scenarios, each involving twenty consecutive lifting tasks. Eight machine learning classifiers were tested utilizing two validation strategies, with the Gradient Boost Tree algorithm achieving the highest accuracy and an Area under the ROC Curve of 91.2% and 94.5%, respectively. Additionally, feature importance analysis was conducted to identify the most influential sway parameters and directions. Results: The results indicate that the combination of sway metrics and the Gradient Boost model offers a feasible approach for predicting biomechanical risks in load lifting. Conclusions: Further studies with a broader participant pool and varied lifting conditions could enhance the applicability of this method in occupational ergonomics. Keywords: biomechanical risk assessment; machine learning; physical ergonomics; postural sway; Revise

FAQs

Why is assessing biomechanical risk in load lifting important?

Long-term work-related musculoskeletal disorders are driven by factors like lifting duration, intensity, and repetitive load. Quantifying biomechanical risk helps reduce back-related injury risk.

How does the study collect postural sway data?

An inertial measurement unit (IMU) is placed at the lumbar region and captures acceleration signals in anteroposterior and mediolateral directions, aligned with the body’s center of mass.

Which machine learning approach performed best and what does it achieve?

Eight classifiers were tested, and the Gradient Boost Tree model achieved the highest accuracy and strong ROC-AUC performance under two validation strategies.

Aug 2, 2026

Upload by Oliver