Self-Organizing Maps for Translating Health Care Knowledge: A Case Study in Diabetes Management
Chronic Disease Management (CDM) is an important area of health care where Health Knowledge Management can provide substantial benefits. A web-based chronic disease management service, called cdmNet, is accumulating detailed data on CDM as it is being rolled out across Australia. This paper presents the application of unsupervised neural networks to cdmNet data to: (1) identify interesting patterns in diabetes data; and (2) assist diabetes related policy-making at different levels. The work is distinct from existing research in: (1) the data; (2) the objectives; and (3) the techniques used. The data represents the diabetes population across the entire primary care sector. The objectives include diabetes related decision and policy making at different levels. The pattern recognition techniques combine a traditional approach to data mining, involving the Self-Organizing Map (SOM), with an extension to include the Growing Self-Organizing Map (GSOM).
KeywordsSpread Factor Chronic Disease Management Business Intelligence Medical Feature Chronic Care Model
Unable to display preview. Download preview PDF.
- 1.WHO Report on Chronic Disease (2005), http://www.slideshare.net/Hfoodsupplements/who-report-on-chronic-disease (accessed: May 12, 2007)
- 2.National chronic disease strategy (2006), http://www.health.gov.au/internet/main/publishing.nsf/content/7E7E9140A3D3A3BCCA257140007AB32B/File/stratal3.pdf (accessed: March 04, 2010)
- 3.Diabetes statistics (2007), http://www.diabetes.org/diabetes-basics/diabetes-statistics/ (accessed: November 12, 2010)
- 4.Diabetes australian facts (2008), http://www.aihw.gov.au/publication-detail/?id=6442468075 (accessed: January 10, 2010)
- 5.Quick facts (2010), http://www.australiandiabetescouncil.com/About-Diabetes/Quick-facts.aspx (accessed: October 12, 2010)
- 6.Viscovery (2010), http://www.viscovery.net/publications/books
- 8.Anderson, G., Wilson, K.: Chronic disease in california: Facts and figures (2006), http://www.chcf.org/~/media/MEDIA%20LIBRARY%20Files/PDF/C/PDF%20ChronicDiseaseFactsFigures06.pdf
- 10.Georgeff, M.: Cdm-net:a broadband health network for transforming chronic disease management. Final report, Precedence Health Care (2010)Google Scholar
- 11.Hsu, W., Lee, M., Liu, B., Ling, T.: Exploration mining in diabetic patients databases: findings and conclusions. In: Proceedings of the Sixth ACM SIGKDD Conf. on KD and DM, p. 436. ACM (2000)Google Scholar
- 13.Huang, Y., McCullagh, P., Black, N.: Feature selection via supervised model construction. In: Fourth IEEE International Conference on Data Mining, ICDM 2004, pp. 411–414. IEEE (2005)Google Scholar
- 16.Porter, T., Green, B.: Identifying Diabetic Patients: A Data Mining Approach. In: Proceedings of the 15th Americas Conference on Information Systems (AMCIS), San Francisco, California (2009)Google Scholar