본문 바로가기
KERD
논문KCI 등재

A Hybrid Bayesian Network Model for Predicting Breast Cancer Prognosis

한태화|박래웅|최종필
대한의료정보학회 (The Korean Society Of Medical Informatics)|2009.03|Healthcare Informatics Research|제15권 제1호|pp. 49-57

2

조회

0

다운로드

4

인용

0

참고문헌

키워드

Neural NetworkBreast CancerPrognosisBayesian Network

초록

Objective: Breast cancer is one of the most common cancers affecting women. Both physicians and patients have concerned about breast cancer survivability. Many researchers have studied the breast cancer survivability applying artificial nerural network model (ANN). Usually ANN model outperformed in classification of breast cancer survivability than other models such as logistic regression, Bayesian network (BN), or decision tree models. However, physicians in the fields hesitate to use ANN model, because ANN is a black-box model, and hard to explain the classification result to patients. In this study, we proposed a hybrid model with a degree of the accuracy and interpretation by combining the ANN for accuracy and BN for interpretation. Methods: We developed an artificial neural network, a Bayesian network, and a hybrid Bayesian network model to predict breast cancer prognosis. The hybrid model combined the artificial neural network and the Bayesian network to obtain a good estimation of prognosis as well as a good explanation of the results. The National Cancer Institute’s SEER program public-use data (1973-2003) were used to construct and evaluate the proposed models. Nine variables, which are clinically acceptable, were selected for input to the proposed models’ nodes. A confidence value of the neural network served as an additional input node to the hybrid Bayesian network model. Ten iterations of random subsampling were performed to evaluate performance of the models. Results: The hybrid BN model achieved the highest area under the curve value of 0.935, whereas the corresponding values of the neural network and Bayesian network were 0.930 and 0.813, respectively. The neural network model achieved the highest prediction accuracy of 88.8% with a sensitivity of 93.7% and a specificity of 85.4%. The hybrid Bayesian network model achieved a prediction accuracy of 87.2% with a sensitivity of 93.3% and a specificity of 83.1%. The results of the hybrid Bayesian network model were very similar to the neural network model. Conclusion: In the experiments, the hybrid model and the ANN model outperformed the Bayesian network model. The proposed hybrid BN model for breast cancer prognosis predictin may be useful for clinicians in the medical fields, as the model provides both high degree of performance inherited from ANN and good explanation power from BN.

Abstract

Objective: Breast cancer is one of the most common cancers affecting women. Both physicians and patients have concerned about breast cancer survivability. Many researchers have studied the breast cancer survivability applying artificial nerural network model (ANN). Usually ANN model outperformed in classification of breast cancer survivability than other models such as logistic regression, Bayesian network (BN), or decision tree models. However, physicians in the fields hesitate to use ANN model, because ANN is a black-box model, and hard to explain the classification result to patients. In this study, we proposed a hybrid model with a degree of the accuracy and interpretation by combining the ANN for accuracy and BN for interpretation. Methods: We developed an artificial neural network, a Bayesian network, and a hybrid Bayesian network model to predict breast cancer prognosis. The hybrid model combined the artificial neural network and the Bayesian network to obtain a good estimation of prognosis as well as a good explanation of the results. The National Cancer Institute’s SEER program public-use data (1973-2003) were used to construct and evaluate the proposed models. Nine variables, which are clinically acceptable, were selected for input to the proposed models’ nodes. A confidence value of the neural network served as an additional input node to the hybrid Bayesian network model. Ten iterations of random subsampling were performed to evaluate performance of the models. Results: The hybrid BN model achieved the highest area under the curve value of 0.935, whereas the corresponding values of the neural network and Bayesian network were 0.930 and 0.813, respectively. The neural network model achieved the highest prediction accuracy of 88.8% with a sensitivity of 93.7% and a specificity of 85.4%. The hybrid Bayesian network model achieved a prediction accuracy of 87.2% with a sensitivity of 93.3% and a specificity of 83.1%. The results of the hybrid Bayesian network model were very similar to the neural network model. Conclusion: In the experiments, the hybrid model and the ANN model outperformed the Bayesian network model. The proposed hybrid BN model for breast cancer prognosis predictin may be useful for clinicians in the medical fields, as the model provides both high degree of performance inherited from ANN and good explanation power from BN.

참고문헌 (25)

  1. Hankey BF, "The surveillance, epidemiology, and end results program: a national resource", Cancer Epidemiol Biomarkers Prev, 8(12), pp.1117~1121 (1999)
  2. Minino AM, "Deaths: preliminary data for 2004", Natl Vital Stat Rep, 54(19), pp.1~49 (2006)
  3. Moore DH, "Breast carcinoma etiological factors", Advances in cancer research, 40, pp.189~253 (1983)
  4. Skolnick MH, "Genetic predisposition to breast cancer", Cancer, 70(6), pp.1747~1754 (1992)
  5. Newcomb PA, "Lactation and breast cancer risk", Journal of mammary gland biology and neoplasia, 2(3), pp.311~318 (1997)
  6. Newcomb PA, "Lactation in relation to postmenopausal breast cancer", American journal of epidemiology, 150(2), pp.174~182 (1999)
  7. Brenin DR, "Factors correlating with lymph nodemetastases in patients with T1 breast cancer", Ann Surg Oncol, 8(5), pp.432~437 (2001)
  8. Kinne DW, "Staging and follow-up of breast cancer patients", Cancer, 67(4), pp.1196~1198 (1991)
  9. Kinne DW, "Axillary clearance in operable breast cancer: still a necessity?", Recent results in cancer research Fortschritte der Krebsforschung, 152, pp.161~169 (1998)
  10. Moore MP, "Is axillary lymph node dissection necessary in the routine management of breast cancer?", Yes. Important advances in oncology, pp.245~250 (1996)
  11. Rosen PP, "Prognosis in T2N0M0 stage I breast carcinoma: a 20-year follow-up study", J Clin Oncol, 9(9), pp.1650~1661 (1991)
  12. Fisher B, "Eight-year results of a randomized clinical trial comparing total mastectomy and lumpectomy with orwithout irradiation in the treatment of breast cancer", The New England journal of medicine, 320(13), pp.822~828 (1989)
  13. Galea MH, "The Nottingham Prognostic Index in primary breast cancer", Breast cancer research and treatment, 22(3), pp.207~219 (1992)
  14. McGuire WL, "Prognostic factors for recurrence and survival in human breast cancer", Breast cancer research and treatment, 10(1), pp.5~9 (1987)
  15. Slamon DJ, "Human breast cancer: correlation of relapse and survival with amplification of the HER-2", neu oncogene, 1987, pp.235(4785), 177~182
  16. Tiezzi DG, "HER-2, p53, p21 and hormonal receptors proteins expression as predictive factors of response and prognosis in locally advanced breast cancer treated with neoadjuvant docetaxel plus epirubicin combination", BMC cancer, 7, pp.36~ (2007)
  17. Burke HB, "Artificial neural networks improve the accuracy of cancer survival prediction", Cancer, 79(4), pp.857~862 (1997)
  18. Choi JW, "Development of Breast Cancer Predication Model", Using Neural Networks, 4(1), pp.83~87 (2003)
  19. Sun­Mi Lee, "Comparisons of predictive modeling techniques for breast cancer in Korean women", Healthcare Informatics Research, 14(1), pp.37~44 (2008), 대한의료정보학회
  20. Lundin M, "Artificial neural networks applied to survival prediction in breast cancer", Oncology, 57(4), pp.281~286 (1999)
  21. Delen D, "Predicting breast cancer survivability: a comparison of three data mining methods", Artificial intelligence in medicine, 34(2), pp.113~127 (2005)
  22. Berner ES, "Clinical Decision Support Systems" (2007), Springer
  23. Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer. gov) Public-Use Data (1973-2003), "SEER Cancer Statistics Review", National Cancer Institute, Surveillance Research Program, Cancer Statistics Branch, ~ (2006)
  24. Rumelhart DE, "arallel distributed processing: explorations in themicrostructure of cognition vol. 2: psychological and biological models" (1986), MIT Press
  25. Jensen FV, "An Introduction to Bayesian Network" (1997), Springer Verlag

관련 논문AI추천

Comparisons of predictive modeling techniques for breast cancer in Korean womenSun­Mi Lee대한의료정보학회 · Healthcare Informatics Research2008 · 제14권 제1호Nomogram of Naive Bayesian Model for Recurrence Prediction of Breast Cancer김구상, 김우재, 박래웅대한의료정보학회 · Healthcare Informatics Research2016 · 제22권 제2호Stacking Ensemble Technique for Classifying Breast Cancer권현진, 박진혁, 이영호대한의료정보학회 · Healthcare Informatics Research2019 · 제25권 제4호Predicting Breast Cancer Survivability: Comparison of Five Data Mining TechniquesArihito Endo, Hiroshi Tanaka, Shibata Takeo대한의료정보학회 · Healthcare Informatics Research2007 · 제13권 제2호딥러닝 기반 앙상블을 이용한 유방암 분류임동훈, 정광모, 최도연한국보건정보통계학회 · 보건정보통계학회지2018 · 제43권 제2호유방암 분류를 위한 전이학습 기반 DNN-SVM 하이브리드 모형김영순, 백범수, 임동훈, 조귀래한국컴퓨터정보학회 · 한국컴퓨터정보학회논문지2023 · 제28권 제11호

이 간행물의 다른 논문 (15건)

Clinical Terminologies: A Solution for Semantic InteroperabilityNick Hardiker, 박현애pp. 1-11Prediction of Hospital Charges for the Cancer Patients withData Mining Techniques정석훈, 서용무, 강진오pp. 13-23건강증진센터를 위한 PDA 활용 모델 개발이학종, 김주영, 한순화, 이민호, 하규섭, 정진엽pp. 25-30건강정보 소비자 용어 시스템 구축을 위한 기반연구김정은, 신윤주pp. 31-40RSSI 방식을 이용한 RFID 기반 환자 위치 모니터링 시스템 설계에 관한 연구김도성, 김정채, 유선국, 김승호pp. 41-48
원문 보기