A Mobile Application to Measure Trunk Flexion Angles in Lifting Tasks

Abstract

Occupational Abstract

The aim of this project was to develop and test an application capable of quickly and repeatedly measuring trunk flexion angles during sagittal plane lifting tasks. The developed application uses the built-in accelerometer in mobile devices to approximate trunk flexion angle, as the user follows an operator as they perform a lift. A black line is superimposed over the camera feed, allowing the user to approximate the angle of inclination of a line connecting the operator’s seventh cervical and first sacral vertebrae—thereby estimating the trunk flexion angle. The magnitude of this angle and its velocity have been linked to the development of occupational low back pain; thus the application provides ergonomists a more refined means of screening tasks beyond currently available survey tools.

TECHNICAL ABSTRACT

Background

The majority of quantitative postural analysis tools used in biomechanics laboratories are either infeasible or impractical for applied ergonomic field use. Survey tools do exist but are subjective in nature.

Purpose

To develop an application for handheld mobile devices that can quickly, reliably, and accurately measure the trunk flexion angle in order to afford more detailed and objective ergonomic analyses.

Methods

The application, Trunk Angle Goniometer (TAG), was programed using Xcode (Apple Inc. Cupertino, CA). Sixteen participants measured the trunk angle of lifts in the sagittal plane using TAG installed on an iPad (Apple Inc., Cupertino, CA). To establish the accuracy of the application, comparisons were made to gold standard (manual anatomical landmark digitization) measures of maximum trunk angle, maximum trunk velocity, and the root-mean square (RMS) difference between trunk angle time histories. Precision was also assessed between raters (inter-rater reliability), between trials assessing the same lift (intra-rater reliability) and between trials assessing similar lifts (test-retest reliability).

Results

TAG generally underestimated the true magnitude of trunk flexion by 5$\mathrm{°}$ to 15$\mathrm{°},$ and overestimated flexion velocity by approximately 10$\mathrm{°}/\text{sec}.$ RMS errors were between 8.6$\mathrm{°}$ and 13.4$\mathrm{°}.$ Performance measures showed fair to good test-retest reliability between 0.631 and 0.709. Overall the application had an excellent inter-rater reliability above 0.95 for all measures; however, suffered from low intra-rater reliability (0.381 to 0.520) but these dramatically increased when averages were taken across multiple trials (from 0.739 to 0.838).

Conclusions

TAG performed well for quantifying angles in the sagittal plane. The approach has the added benefit of being able to assess lifting tasks in real time, combined with its relatively cheap cost, the approach shows promise for field-work and assessments.

Keywords:

• Trunk angle goniometer
• trunk flexion
• lifting
• mobile application

Acknowledgements

The authors would like to acknowledge funding from the Natural Science and Engineering Research Council of Canada (NSERC) and the Center of Research Expertise for the Prevention of Musculoskeletal Disorders (CRE-MSD). Dr. Jack P. Callaghan is supported by the Canada Research Chair in Spine Biomechanics and Injury Prevention. Special thanks is given to James Marles for his assistance in digitizing landmarks for this study.

Additional information

Funding

Jack P. Callaghan is a Canada Research Chair in Spine Biomechanics and Injury Prevention. This work was supported by a Natural Sciences and Engineering Council Undergraduate Research Apprenticeship (USRA) Scholarship and Center for Research Excellence for the Prevention of Musculoskeletal Disorders (CRE-MSD).