Using Artificial Neural Network as a Tool for Epidemiological Data Analysis
The importance of socio-economic part of medicine still increases. This study was performed to investigate abilities to use Artificial Neural Network as a tool for epidemiological data analysis. Back-propagation neural networks were simulated on own-written software. The sensitivity analysis results of created models suggest that ANN are able to discover most significant factors for studied output and as a consequence of this could be helpful for medical policy makers in decision process.
KeywordsArtificial Neural Network Heart Attack Sensitivity Analysis Result Total Health Care Cost Good Architecture
Unable to display preview. Download preview PDF.
- 1.Osowski S. (1996) Neural network in algorithmic depiction. Scientific-Technical Publishing House, Warsaw (In Polish)Google Scholar
- 2.De Laurentis A., et al. (1999) A prognostic model that makes quantitative estimates of probability o` relapse for breast cancer patients. Clinical Cancer Research vol. 5, 4133–4139Google Scholar
- 3.Plate T., Band P., Bert J., Grace J., (1997) A comparison between neural networks and other statistical techniques for modelling the relationships between tobacco and alcohol and cancer., Advances in Neural Information Processing, MIT Press, 967–973Google Scholar
- 6.Morrison J.R., Johnson J.D., Barnes J.H., Summers K., Szeinbach S.L., (1997) Predicting Total Health Care Costs of Medicaid Recipients: An Artificial Neural System Approach, 40, 191–197Google Scholar