Wearable solution for online assessment of biomechanical load risks

Work-related Musculo Skeletal Disorders (WMSD) are considered the third main reason for disability and early retirement in the U.S. and are widespread in many occupations. In Italy, only taking into account the years 2009-2010, it is estimated an exponential increasing in the number of WMSD reports. In particular a 159.7 % increment has been reported compared to the 2006 statistics. In this context, it is clear how important correctly diagnosing this kind of pathology is becoming. Traditional methods for WMDS assessment are based on observational techniques, in which experts manually segment, label and evaluate movements with the help of pro-forma sheets. Since these methods are currently based on visual inspection and subjective judgment, they could benefit from objective measurements in terms of both reliability and repeatability. Moreover an automatic tool for ergonomics assessment would vastly reduce the time that an expert needs to carry out the same assessment manually. In this context we propose a novel wearable wireless system capable of assessing the muscular efforts and postures of the human upper limb for WMSDs diagnosis. The system, being non-obstructive, can be used to monitor workers in ecologic environment while they are carrying on their everyday tasks. Different methods assessing different factors lead to different risk evaluations. For this reason using more than one method at the same time can help prioritize interventions and ensure a more thorough evaluation of risk factors. The proposed system provides real-time assessment obtained according to two of the most common indexes for the analysis of risk factors on workplaces: the Rapid Upper Limb Assessment (RULA) and the Strain Index (SI). The manual assessment of multiple methods would lead to an unacceptably high cost in terms of time and money for practitioners. The system exploits inertial measurement units (IMUs) to reconstruct the upper limb posture, modeled as a 7 degrees of freedom (DoF) kinematic chain. As far as muscular efforts are concerned, surface EMG sensors are used to assess forearm flexor muscles strain. A case study application of the system is presented showing the results of a first data collection campaign regarding super-market cashiers during everyday real-life operations. The system assessment has been compared using two human evaluators as ground truth. The system RULA score accuracy is 94.79%, while the SI score shows an accuracy of 50%. Further investigation will be carried on taking also into account the subjects' perceived effort using self-assessment sheets.