Comparison of trunk parameters and its application in an automated REBA system: A pilot study

A high prevalence of work-related musculoskeletal disorders (WMSDs) among the workers from different industries has been reported. REBA ergonomic quick assessment tool is widely used to evaluate the WMSDs risks of a variety of tasks. Previous studies analysed the REBA scores of just a few specific frames due to the manual evaluation done by the experts. Moreover, the postural angles, which are the input of the REBA assessment, are previously analysed using observation, a goniometer, or video processing tools. These methods might cause errors in obtaining the angles. Recently, the Xsens MVN Inertial Measure Units (IMUs) and an automated REBA system were used to improve accuracy and efficiency. However, the trunk angles obtained by the system were underestimated, which reduced the trunk REBA score and the overall REBA score. This study aimed to identify the best-represented trunk angle parameter in the Xsens MVN and to develop an automated REBA system which used the identified trunk angle parameter. The data samples used for analysis were obtained from the databases from a clinical trial study about patient transfer tasks. The subjects were put on with Xsens IMUs and carried out tasks which involved trunk flexion and lateral bending. Five frames were randomly chosen for trunk flexion analysis and two frames for lateral bending. Three types of parameters related to the trunk angles were obtained from the Xsens data and these angles were compared to the angles identified from the photo samples. An automated REBA system was created using the MATLAB platform and the REBA scores were compared to the scores generated from the existing automated REBA system. Vertical-T8 angles have the lowest difference (1.2°±1.0° for flexion and 6.2°±0.0° for lateral bending) from the trunk angles identified in the photos. The existing REBA system underestimated the trunk angles which caused an underestimation of the trunk REBA score and overall REBA score.