PAKDD 2013: Trends and Applications in Knowledge Discovery and Data Mining pp 143-154 | Cite as
On the Application of Multi-class Classification in Physical Therapy Recommendation
Abstract
Recommending optimal rehabilitation intervention for injured workers that would lead to successful return-to-work (RTW) is a challenge for clinicians. Currently, the clinicians are unable to identify with complete confidence which intervention is best for a patient and the referral is often made in trial and error fashion. Only 58% recommendations are successful in our dataset. We aim to develop an interpretable decision support system using machine learning to assist the clinicians. We use various re-sampling techniques to tackle the multi-class imbalance and class overlap problem in real world application data. The final model has shown promising potential in classification compared to human baseline and has been integrated into a web-based decision-support tool that requires additional validation in a clinical sample.
Keywords
multi-class imbalance re-sampling clinical decision-support rule-based learningPreview
Unable to display preview. Download preview PDF.
References
- 1.Two modifications of cnn. IEEE Transactions on Systems, Man and Cybernetics SMC-6(11), 769–772 (November 1976)Google Scholar
- 2.CMAR: accurate and efficient classification based on multiple class-association rules (2001)Google Scholar
- 3.Martin, B.I., Deyo, R.A., Mirza, S.K., Turner, J.A., Comstock, B.A., Hollingworth, W., Sullivan, S.: Expenditures and health status among adults with back and neck problems. JAMA 299, 656–664 (2008)CrossRefGoogle Scholar
- 4.Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P.: Smote: synthetic minority over-sampling technique. J. Artif. Int. Res. 16(1), 321–357 (2002)MATHGoogle Scholar
- 5.Murray, C.J., Vos, T., Lozano, R., Naghavi, M., Flaxman, A., Michaud, C., et al.: Disability-adjusted life years (dalys) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the global burden of disease study 2010. Lancet 380(9859), 2197–2223 (2012)CrossRefGoogle Scholar
- 6.Cohen, W.W.: Fast effective rule induction. In: Proceedings of the Twelfth International Conference on Machine Learning, pp. 115–123 (1995)Google Scholar
- 7.Hadler, N.M.: Occupational musculoskeletal disorders, 3rd edn. Lippincott Williams & Wilkins, Philadelphia (2005)CrossRefGoogle Scholar
- 8.Lane, R., Desjardins, S.: Canada. population and public health branch. Strategic policy directorate. Policy research division. Economic burden of illness in Canada [Ottawa], Health Canada (2002)Google Scholar
- 9.Laurikkala, J.: Improving Identification of Difficult Small Classes by Balancing Class Distribution. In: Quaglini, S., Barahona, P., Andreassen, S. (eds.) AIME 2001. LNCS (LNAI), vol. 2101, pp. 63–66. Springer, Heidelberg (2001)CrossRefGoogle Scholar
- 10.Ross Quinlan, J.: C4.5: programs for machine learning. Morgan Kaufmann Publishers Inc., San Francisco (1993)Google Scholar
- 11.Wilson, D.L.: Asymptotic properties of nearest neighbor rules using edited data. IEEE Transactions on Systems, Man and Cybernetics 2(3), 408–421 (1972)MATHCrossRefGoogle Scholar
- 12.Zaïane, O.R., Antonie, M.-L.: Classifying text documents by associating terms with text categories. In: Proceedings of the 13th Australasian Database Conference, ADC 2002, Darlinghurst, Australia, vol. 5, pp. 215–222. Australian Computer Society, Inc. (2002)Google Scholar