A model for predicting the risk of musculoskeletal disorders among computer professionals

Objective. This study aimed to develop a model for predicting the risk of musculoskeletal disorders among computer professionals. Materials and methods. A preliminary study with a modified Nordic musculoskeletal questionnaire was conducted to identify the risk in different body parts of the professionals during their work. A discrete postural evaluation of the dynamic postures involved in the work was assessed using rapid upper limb assessment. Postural, physiological and work-related factors were considered as attributes of the model. The model was developed using various machine learning algorithms, and was then tested and validated. Results. The postural factor of the computer professionals was found to be significantly (p < 0.01) correlated with the musculoskeletal disorders. Results of the logistic regression analysis showed that physiological and work-related factors were also significantly (p < 0.05) associated with musculoskeletal disorders. The Random Forest algorithm and Naïve Bayes Classifier predicted the risk of musculoskeletal disorders with the highest accuracy (81.25%). Conclusion. Postural, physiological and work-related factors contribute to the development of musculoskeletal disorders. The Random Forest algorithm or Naïve Bayes Classifier model developed based on these factors could be used to accurately predict the risk of musculoskeletal disorders among computer professionals at any instance of time, during their work.