Wearable Sensors and Machine Learning for the Assessment of Biomechanical Risk in Lifting Tasks

Wearable Sensors and Machine Learning: a Technological Approach to Biomechanical Risk Assessment in Lifting Tasks

Lifting loads can cause work-related musculoskeletal disorders. The National Institute for Occupational Safety and Health (NIOSH) established a methodology for assessing lifting actions by means of a quantitative method based on intensity, duration, frequency, and other geometrical characteristics of lifting. Body-worn inertial sensor technology provides a number of opportunities to advance the safety and health of workers engaged in physical work. Motion-tracking systems together with Machine learning (ML) algorithms are used in the ergonomic field for biomechanical risk assessment by means of data acquired by wearable inertial systems. The investigators posed the question whether it is possible to classify lifting tasks belonging to different risk classes according to the value of LI using a machine learning approach by means of features extracted from raw signals. Aim of this study was to develop and validate, through ML algorithms, a non-invasive detection system of kinetic-kinematic parameters using IMU and EMG sensors, for the ergonomic assessment of the risk associated with a load lifting activity.

Study Overview

• Status: Completed
• Conditions: Wearable Devices
• Intervention / Treatment: Device: wearable device

Detailed Description

The study envisages the voluntary enrollment of healthy subjects, referring to treatment clinics for work-related pathologies (excluding subjects aged <18 or > 65 years, and those with musculoskeletal pathologies or other disabling pathologies in progress), to carry out two repeated lifting tests. The two tests are set up to correspond respectively to the two NIOSH risk classes (LI<1, NO RISK; and LI>1, RISK). The IMU sensors provide wirelessly a series of data from which it is intended to extract a number of features (feature extraction) that have a high predictive power, through the digital signal processing technique using dedicated software (i.e. Matlab, SPSS). In a second step, data obtained from EMG sensors will be added to the analysis. Among the different artificial intelligence algorithms, the investigator will look for those most able to discriminate the various risk classes on the basis of the parameters extracted from the signals detected during the motor task.

• Study Type: Observational
• Enrollment (Actual): 41

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 65 years (Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

healthy volunteer referring to treatment clinics for work-related pathologies

Description

Inclusion Criteria:

• healthy subjects

Exclusion Criteria:

• subjects with musculoskeletal pathologies or other disabling pathologies in progress

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Validation of the proposed strategy to assess the risk of lifting activities, according to RNLE	accuracy degree and AucRoc	first year

Collaborators and Investigators

• Sponsor: Istituti Clinici Scientifici Maugeri SpA
• Investigators: Principal Investigator: Edda Capodaglio, PhD, ICS Maugeri IRCCS

Publications and helpful links

General Publications

• Donisi L, Cesarelli G, Capodaglio E, Panigazzi M, D'Addio G, Cesarelli M, Amato F. A Logistic Regression Model for Biomechanical Risk Classification in Lifting Tasks. Diagnostics (Basel). 2022 Oct 29;12(11):2624. doi: 10.3390/diagnostics12112624.
• Donisi L, Cesarelli G, Pisani N, Ponsiglione AM, Ricciardi C, Capodaglio E. Wearable Sensors and Artificial Intelligence for Physical Ergonomics: A Systematic Review of Literature. Diagnostics (Basel). 2022 Dec 5;12(12):3048. doi: 10.3390/diagnostics12123048.
• Donisi L, Capodaglio EM, Amitrano F, Cesarelli G, Pagano G, D'Addio G. A multiple linear regression approach to extimate lifted load from features extracted from inertial data. G Ital Med Lav Ergon. 2021 Dec;43(4):373-378.
• Donisi L, Cesarelli G, Coccia A, Panigazzi M, Capodaglio EM, D'Addio G. Work-Related Risk Assessment According to the Revised NIOSH Lifting Equation: A Preliminary Study Using a Wearable Inertial Sensor and Machine Learning. Sensors (Basel). 2021 Apr 7;21(8):2593. doi: 10.3390/s21082593.

Study record dates

Study Major Dates

• Study Start (Actual): October 7, 2010
• Primary Completion (Actual): January 24, 2022
• Study Completion (Actual): May 6, 2022

Study Registration Dates

• First Submitted: March 2, 2023
• First Submitted That Met QC Criteria: March 17, 2023
• First Posted (Actual): March 21, 2023

Study Record Updates

• Last Update Posted (Actual): December 28, 2023
• Last Update Submitted That Met QC Criteria: December 21, 2023
• Last Verified: December 1, 2023

More Information

Terms related to this study

• Keywords: Ergonomics; Machine learning; Risk prediction; Accelerometers; Lifting; inertial measurement units; Exposure assessment; Work-related musculoskeletal disorders; Digital signal processing
• Other Study ID Numbers: 2475

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No