Semi-Supervised Learning Based Prediction of Musculoskeletal Disorder Risk

Document Type: Technical note

Authors

1 Mechanical Engineering Department, National Institute of Technology, Kurukshetra 136119, Haryana, INDIA

2 Civil Engineering Department, National Institute of Technology, Kurukshetra 136119, Haryana, INDIA

Abstract

This study explores a semi-supervised classification approach using random forest as a base classifier to classify the low-back disorders (LBDs) risk associated with the industrial jobs. Semi-supervised classification approach uses unlabeled data together with the small number of labelled data to create a better classifier. The results obtained by the proposed approach are compared with those obtained by a backpropagation neural network. Comparison indicates an improved performance by the semi-supervised approach over the random forest classifier as well as neural network approach. Highest classification accuracy of 78.20% was achieved by the used semi-supervised approach with random forest as base classifier in comparison to an accuracy of 72.4% and 74.7% obtained by random forest and back propagation neural network approaches respectively. Thus results suggest that the proposed approach can successfully classify jobs into the low and high risk categories of low-back disorders based on lifting task characteristics.

Keywords

Main Subjects


[1] Breiman L. (1996), Bagging predictors; Machine Learning 26; 123-140.
[2] Breiman L. (1999), Random forests-Random Features; Technical Report 567; Statistics Department,
University of California, Berkeley, http://ftp.stat.berkeley.edu/pub/users/breimanf,4 July 2007.
[3] Bureau of labour statistics (1996), Characteristics of injuries and illnesses resulting in absences from
work, 1994 Washington, DC: US Department of Labour; Bureau of Labour Statistics, USDL; 96–163.
[4] Dempsey P.G., Ayoub M.M., Westfall P.H. (1995), The NIOSH lifting equations: a closer look. In:
Bitner A.C. and Champney P.C. (Ed); Advance in Industrial Ergonomics and Safety VII. Taylor and
Francis, Bristol, PA; 705–712.

[5] Dempsey P.G., Westfall P.H. (1997), Developing explicit risk models for predicting low-back
disability: a statistical perspective; International Journal of Industrial Ergonomics 19; 483–497.
[6] Driessens K., Reutemann P., Pfahringer B., Leschi C.(2006), Using Weighted Nearest Neighbour to
Benefit from Unlabeled Data.In:Wee Keong Ng,Masaru Kitsuregawa, Jianzhong Li, and Kuiyu Chang,
(Ed); Advances in Knowledge Discovery and Data Mining, 10th Pacific-AsiaConference, PAKDD
2006, 3918 of LNCS; 60-69.
[7] Feller W. (1968), An Introduction to Probability Theory and Its Application; third edition, Vol. 1,
Wiley; New York.
[8] Frymoyer J.W., Cats-Baril W.L. (1991), An overview of the incidence and costs of low-back pain;
Orthopaedic Clinics of North America 22; 262–271.
[9] Guo H., Tanaka S., Cameron L., Seligman P., Behrens V., Ger J. (1995), Back pain among workers in
the United States: national estimates and workers at high risks; American Journal of Industrial
Medicine 28; 591–602.
[10] Karwowski W., Zurada J., Marras W.S., Gaddie P. (1994), A prototype of the artificial neural networkbased
system for classification of industrial jobs with respect to risk of low back disorders. In:
Aghazadeh F. (Ed); Advances in Industrial Ergonomics and Safety VI, Taylor & Francis, Bristol, PA;
19–22.
[11] Killough M.K., Crumpton L.L., Calvert A., Bowden R. (1995), An investigation using neural networks
to identify the presence of carpal tunnel syndrome; 4th Industrial Engineering Research Conference
Proceedings, IIE, Norcross, GA; 659–667.
[12] Marras W.S., Lavender S.A., Leurgans S., Sudhakar L.R., Allread W.G., Fathallah F., Ferguson S.
(1993), The role of dynamic three dimensional trunk motion in occupationally-related low back
disorders; Spine 18; 617–628.
[13] Pal M., Mather P.M. (2003), An assessment of the effectiveness of decision tree methods for land
cover classification; Remote Sensing of Environment 86; 554–565.
[14] Quinlan J.R. (1993), C4.5: Programs for Machine Learning; Morgan Kaufmann, San Mateo.
[15] Seeger M. (2002), Learning with labeled and unlabeled data; Technical report, Institute for Adaptive
and Neural Computation, University of Edinburgh,
http://www.kyb.tuebingen.mpg.de/bs/people/seeger/papers/review.pdf, 13 May 2007.
[16] Tanaka S., Wild D., Seligman P., Halperin W., Behrens V., Putz-Anderson V. (1995), Prevalence and
work-relatedness of self-reported carpal tunnel syndrome among US workers: Analyses of the
occupational health supplement data to the 1988 National Health Interview Survey; American Journal
of Industrial Medicine 27; 451–470.
[17] Webster B.S., Snook S.H. (1994), The cost of compensable upper extremity cumulative trauma
disorders; Journal of Occupational and Environmental Medicine 36; 713–717.
[18] Zhu X. (2006), Semi-supervised learning literature survey; Technical Report 1530, Department of
Computer Sciences, University of Wisconsin-Madison,
http://www.pages.cs.wisc.edu/~jerryzhu/pub/ssl_survey.pdf, 8 May 2007.
[19] Zurada J., Karwowski W., Marras W.S. (1997), A neural network-based system for classification of
industrial jobs with respect to risk of low back disorders due to work-place design; Applied
Ergonomics 28; 49–58.