Design And Development Of Musculoskeletal Disorders Risk Assessment System Based On Posture Recognition

With the development of science and technology,the traditional occupational disorders related to harmful gases,vibration and other environmental factors have been gradually controlled.However,musculoskeletal disorders（MSDs）have become the main occupational disorders affecting workers’occupational health due to its non-representation in the early stage and suddenness in the later stage.MSDs is a kind of occupational disorders caused by poor working posture and high-intensity work load.The symptoms of MSDs are mainly pain in legs,wrists,lower back,and neck.MSDs will not only directly affect the health of workers,but also cause huge losses to the social and economic.Therefore,evaluating the MSDs caused by workers’working posture and taking scientific corrective measures can effectively reduce the incidence of MSDs.However,the existing MSDs evaluation methods are time-consuming,labor-intensive and expensive in use.Therefore,based on the occupational health of worker,combined with the knowledge of ergonomics,musculoskeletal model,deep learning and image recognition,a risk assessment system of MSDs based on posture recognition is established.In order to accurately and quickly assess the potential risk of musculoskeletal disorders in the workplace and improve the health of workers.The main contents and innovations of this research are as follows:1)The theoretical feasibility of MSDs risk assessment based on work posture was studied.Based on the human musculoskeletal anatomy,the human skeleton joint point model is defined,and the feasibility of MSDs risk assessment is analyzed,which provides a theoretical basis for MSDs assessment.2)A MSDs risk assessment system based on human skeleton recognition algorithm is proposed.And completed the theoretical design and practical development of the system,named Quick Capture Evaluation System（QCES）.The QCES system combined with artificial intelligence algorithm solves the problem of MSDs risk assessment.3)The feasibility and reliability of QCES system were verified through experiments.The consistency of QCES system and motion capture system in limb angle recognition is compared.The consistency of QCES system and ergonomic experts in MSDs risk assessment score is compared.The experiment results show that,the average Root Mean Square Error（RMSE）was 4.77,and the average Spearman’s rho（ρ）correlation coefficient was 0.915.in terms of evaluation score,the consistency coefficient（P₀）was 0.952 and kappa coefficient was 0.738.It is proved that QCES system can not only accurately identify and analyze the working posture,but also accurately assess the MSDs risk level of current working posture.4)Finally,the QCES system was applied to practice,analyzed and evaluated the working posture of workers in the actual complex and severely occluded working environment.The results were analyzed,and the corresponding improvement suggestions were put forward.