Lifting activity assessment using surface electromyographic features and neural networks

The surface electromyographic (sEMG) data of 12 trunk muscles of 10 workers during the execution oflifting tasks using three lifting indices (LI) were recorded. The aims of this work were to: 1) identify themost sensitive trunk muscles with respect to changes in lifting conditions based on the selected sEMGfeatures and 2) test whether machine-learning techniques (artificial neural networks) used for mappingtime and frequency sEMG features on LI levels can improve the biomechanical risk assessment. Theresults show that the erector spinae longissimus is the trunk muscle for which every sEMG feature cansignificantly discriminate each pair of LI. Furthermore, only when using multi-domain features (time andfrequency) a more complex artificial neural network can lead to an improved biomechanical risk clas-sification related to lifting tasks