Many work activities can imply a biomechanical overload. Among these activities, lifting loads may determine work-related musculoskeletal disorders. To limit injuries, the National Institute for Occupational Safety and Health (NIOSH) proposed a methodology to assess biomechanical risk in lifting tasks through an equation based on intensity, duration, frequency and other geometrical characteristics of lifting tasks. In this work, we explored the feasibility of tree-based machine learning algorithms to classify biomechanical risk according to the Revised NIOSH lifting equation). Electromyography signals acquired from the biceps and sternum acceleration signals collected during lifting loads were registered using a wearable sensor (BITalino (r)evolution) worn by 5 healthy young subjects. Electromyography and acceleration signals were segmented as to extract the region of interest related to the lifting actions and, for each region of interest, several time and frequency domain features were extracted. Interesting results were obtained in terms of evaluation metrics for a binary risk/no- risk classification. In conclusion, this work indicates the proposed combination of features and machine learning algorithms represents a valid approach to automatically classify risk activities according to the Revised NIOSH lifting equation. Future investigation on enriched study populations could confirm the capabilities of this methodology to automatically classify potential risky activities.
Machine Learning and Biosignals are able to discriminate biomechanical risk classes according to the Revised NIOSH Lifting Equation
Donisi, L;
2022
Abstract
Many work activities can imply a biomechanical overload. Among these activities, lifting loads may determine work-related musculoskeletal disorders. To limit injuries, the National Institute for Occupational Safety and Health (NIOSH) proposed a methodology to assess biomechanical risk in lifting tasks through an equation based on intensity, duration, frequency and other geometrical characteristics of lifting tasks. In this work, we explored the feasibility of tree-based machine learning algorithms to classify biomechanical risk according to the Revised NIOSH lifting equation). Electromyography signals acquired from the biceps and sternum acceleration signals collected during lifting loads were registered using a wearable sensor (BITalino (r)evolution) worn by 5 healthy young subjects. Electromyography and acceleration signals were segmented as to extract the region of interest related to the lifting actions and, for each region of interest, several time and frequency domain features were extracted. Interesting results were obtained in terms of evaluation metrics for a binary risk/no- risk classification. In conclusion, this work indicates the proposed combination of features and machine learning algorithms represents a valid approach to automatically classify risk activities according to the Revised NIOSH lifting equation. Future investigation on enriched study populations could confirm the capabilities of this methodology to automatically classify potential risky activities.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.