PREDICCIÓN DEL RIESGO DE LESIONES MÚSCULO‐ESQUELÉTICAS EN EL LEVANTAMIENTO DE CARGAS MEDIANTE REDES NEURONALES

Abstract

This paper describes a new neural network model applied to classifying the risk of low back disorders presented by certain lifting jobs. The neural network obtained can be used by the ergonomist as a diagnostics system, enabling jobs to be classified into two categories (lowrisk and high-risk) according to the associated likelihood of causing low back disorders. This system provides a higher proportion of correct classifications (81.6%) than other previous models. The development process of neural networks for classification problems and the influence of network architecture on its prediction and generalisation capabilities are analysed. So it is the phenomenon of overfitting and its relationship to the number of network connections and the size of the training data set. The new approach uses complex architecture networks and procedures for avoiding overfitting. It is compared with previous works and its results and advantages over them are discussed. The neural network obtained has been implemented in a software application focused on risk analysis and prevention of the injuries caused by tasks involving manual lifting in the industrial environment. Keywords: artificial neural networks; low-back disorders; ergonomic assessment