Predicting relative and working heart rates of bricklaying jobs using neural network

BACKGROUND: Many ways have been adopted in measuring workers’, responses to manual tasks in order to appraise the incompatibility of work demands to the capabilities of the workers. Heart rate is commonly used to estimate the energy expenditure or physical strain in physically demanding job. OBJECTIVE: The main purpose of this study was to build a prediction model using the neural network to reflect the effects of age, body height, body mass, and resting heart rate on the working heart rate (HWorking) and % relative heart rate (%RHR) of bricklayers. METHODS: A neural network in SPSS 16.0 was applied to identify the importance of the inputs (age, body height, body mass, resting heart rate) in predicting the outputs (working heart rate and % RHR) of a function. RESULTS: The results show that the mean % relative heart rate (RHR) was 57.4%. The mean working heart rate was 120.8 bpm and that of resting heart rate was 68.6 bpm. It was also shown that the neural network could be trained to predict HWorking and % RHR. This also demonstrates that there is a non-linear relationship between the age, body height, body mass, resting heart rate, working heart rate and % RHR. The neural network results for the HWorking and % RHR were dominated by resting heart rate, followed by body mass, age and body height in that order. The predicted values of % RHR and HWorking did not differ significantly from the actual values though the relationships were non-linear. CONCLUSIONS: The neural network might be used to predict the % RHR and HWorking of bricklayers in Nigeria given age, body height, body mass and resting heart rate.