An Application of Artificial Immune Recognition System for Prediction of Diabetes Following Gestational Diabetes
- 248 Downloads
Diabetes mellitus (DM) is a disease prevalent in population and is not easily perceived in its initial stage but may sway a patient very seriously in later stage. In accordance with the estimation of World Health Organization (WHO), there will be 370 million diabetics which are 5.4% of the global people in 2030, so it becomes more and more important to predict whether a pregnant woman has or is likely to acquire diabetes. This study is conducted with the use of the machine learning—Artificial Immune Recognition System (AIRS)—to assist doctors in predicting pregnant women who have premonition of type 2 diabetes. AIRS is proposed by Andrew Watkins in 2001 and it makes use of the metaphor of the vertebrate immune system to recognize antigens, select clone, and memorize cells. Additionally, AIRS includes a mechanism, limited resource, to restrain the number of memory cells from increasing uncontrollably. It has also showed positive results on problems in which it was applied. The objective of this study is to investigate the feasibility in using AIRS to predict gestational diabetes mellitus (GDM) subsequent DM. The dataset of diabetes has imbalanced data, but the overall classification recall could still reach 62.8%, which is better than the traditional method, logistic regression, and the technique which is thought as one of the powerful classification approaches, support vector machines (SVM).
KeywordsArtificial Immune Recognition System (AIRS) Type 2 diabetes Vertebrate immune system Imbalanced data
This work was supported in part by the National Science Council, Taiwan, under grant NSC-98-2221-E-007-071-MY3.
- 2.Seshiah, V., Balaji, V., Balaji, M. S., Panneerselvam, A., Thamizharasi, M., and Arthi, T., Glycemic level at the first visit and prediction of GDM. J. Assoc. Phys. India. 55:630–632, 2007.Google Scholar
- 4.Campos, M. A. A., Reichelt, A. A. J., Façanha, C., Forti, A. C., and Schmidt, M. I., Evaluation of a 1-h 75-g oral glucose tolerance test in the diagnosis of gestational diabetes. Braz. J. Med. Biol. Res. 41(8):684–688, 2008.Google Scholar
- 8.Carr, D. B., Newton, K. M., Utzschneider, K. M., Tong, J., Gerchman, F., Kahn, S. E., and Heckbert, S. R., Modestly elevated glucose levels during pregnancy are associated with a higher risk of future diabetes among women without gestational diabetes mellitus. Diabetes Care. 31(5):1037–1039, 2008.CrossRefGoogle Scholar
- 10.Watkins, A. B. AIRS: A resource limited artificial immune classifier. Master thesis, Mississippi State University, 2001.Google Scholar
- 15.Nasaroui, O. F., Gonzalez, F., and Dasgupta, D., The fuzzy artificial immune system: motivation, basic concepts, and application to clustering and web profiling. International Joint Conference on Fuzzy Systems, pp. 711–717, 2002.Google Scholar
- 17.Goodman, D. E. Jr., Boggess, L. C., and Watkins, A. B., Artificial immune system classification of multiple-class problems. The Proceeding of the Artificial Neural Networks in Engineering, pp. 179–183, 2002.Google Scholar
- 22.Kubat, M., and Matwin, S., Addressing the curse of imbalanced training set: one-sided selection. Proceeding 14th International Conference on Machine Learning (ICML’97) 179–186, 1997.Google Scholar
- 23.Chang, C. C., Lin, C. J. LIBSVM: a library for support vector machines, 2001. Software available at http://www.csie.ntu.edu.tw/∼cjlin/libsvm
- 24.Hsu, C. W., and Lin, C. J., A comparison of methods for multiclass support vector machines. IEEE Trans. Neural Netw. 12(2):415–425, 2002.Google Scholar