HIV 환자 입원 예측 연구를 위한 베이시안 네트워크 활용 = Application of a 3 Bayesian Network to Predict Hospitalization among HIV Adults

다국어 초록 (Multilingual Abstract)

Objective: The purpose of this study was to explore the potential application of a Bayesian network, an emerging data mining technique, in predicting outcomes using large healthcare databases. Methods: The HIV Cost and Services Utilization Study(HCSUS) dataset, consisting of 2,864 HIV positive adults in the US, was used. A total of 35 variables were selected including one output variable defined as more than one hospitalization in six months representing a sub-optimal pattern of healthcare utilization in HIV care. The HUGIN Researcher 6.2 was used to develop a Bayesian network model with two learning algorithms: 1) Necessary Path Condition(NPC) to construct a Bayesian network structure, and 2) Expectation-Maximization(EM) algorithm to estimate parameters. Results: The area under the Receiver Operating Characteristic(ROC) curve was .72. The accuracy of the prediction model was .66. Sensitivity and specificity were .65 and .66, respectively.Conclusion: The Bayesian network showed fair performance in this prediction problem. This study provides researchers new insight into working with large sets of data, which continue to grow in a number of cases and variables. The repeated testing and refinement of the Bayesian network modeling techniques with other health outcomes in large databases is recommended.

참고문헌 (Reference)

1 "Variations in the care of HIV-infected adults in the United States Results from the HIV Cost and Services Utilization Study" 281 (281): 2305-2315, jama1999

2 "Transforming classifier score into accurate multiclass probability estimates Proceedings of the Eighth International Conference on Knowledge Discovery and Data Mining 2002" 694-699, 2002

3 "The EM algorithm for graphical association models with missing data" 19 : 191-201, 1995

4 "Probabilistic reasoning in intelligent systems Networks of plausible inference,San Francisco, CA" Morgan Kaufmann Publishers 1988

5 "Probabilistic networks and expert systems" Springer 1999

6 "Neural networks in clinical medicine," 16 (16): 386-398, 1996

7 "Missing data in nursing research Review of issues and treatment strategies" ;2 (;2): 31-38, 2003

8 "Medical data mining and knowledge discovery." Physica-Verlag 2001

9 "Learning Bayesian networks: The combination of knowledge and statistical dataMSR-TR-94-09 Microsoft Research" 1995

10 "Knowledge discovery in large data sets A primer for data mining applications in health . Nursing Informatics" 139-148, 2000

11 "Introduction to health services" Delmar Publisher 87-109, 1999

12 "Indicators and predictors of health services utilization Introduction to health services" Delmar Publisher Inc 46-70, 1993

13 "Health services utilization models Handbook of health behavior research I" Plenum Press 153-172, 1997

14 "Handling missing data in survey research" 5 (5): 215-238, 1996;

15 "Goodness of fit tests for the multiple logistic regression model Communications in Statistics 1980" 1043-1069,

16 "Equity of access to medical care" 4-27,

17 "Diagnosing community-acquired pneumonia with a Bayesian network" 632-636, 1998;

18 "Conditions for the optimality of the simple Bayesian Classifier Proceeding of the Thirteenth International Conference on Machine Learning 1996" 105-112,

19 "Bayesian networks for knowledge discovery in large datasets Basics for nurse researchers" (4 ((4): 389-399, 2003

20 "Bayesian networks for knowledge discovery Advances in knowledge discovery and data mining The MIT Press" CA: The MIT Press 273-305, 1996

21 "Bayesian networks for data mining" 1 : 79-119, 1997

22 "An introduction to Bayesian networks" UCL Press 1996

23 "Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes" 49 : 1225-1232, 1996

24 "Advances in knowledge discovery and data mining" MIT Press 1996

25 "A comparison of three techniques for rapid model development An application in patient risk-stratification" 443-447, 1996

26 "A Bayesian network for mammography" 106-110, 2000