The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Nomogram comparison conducted by logistic regression and naïve Bayesian classifier using type 2 diabetes mellitus (T2D)

제 2형 당뇨병을 이용한 로지스틱과 베이지안 노모그램 구축 및 비교

  • Received : 2018.04.09
  • Accepted : 2018.06.08
  • Published : 2018.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we fit the logistic regression model and naïve Bayesian classifier model using 11 risk factors to predict the incidence rate probability for type 2 diabetes mellitus. We then introduce how to construct a nomogram that can help people visually understand it. We use data from the 2013-2015 Korean National Health and Nutrition Examination Survey (KNHANES). We take 3 interactions in the logistic regression model to improve the quality of the analysis and facilitate the application of the left-aligned method to the Bayesian nomogram. Finally, we compare the two nomograms and examine their utility. Then we verify the nomogram using the ROC curve.

본 연구에서는 제 2형 당뇨(type 2 diabetes mellitus)의 발병 확률을 예측하기 위해 11가지 위험요인을 가지고 로지스틱 회귀모형과 순수 베이지안 분류기 모형에 적합시킨다. 그런 다음 이를 시각적으로 쉽게 이해하는데 도움을 주는 노모그램 구축 방법을 소개한다. 분석은 2013-2015년 6기 국민건강영양조사 데이터를 가지고 분석하였다. 또 로지스틱 회귀모형에 세 가지 상호작용 항을 넣어 분석의 질을 높이고자 하였고 베이지안 노모그램에 left-aligned 방법을 사용하여 비교하기 쉽게 만들었다. 최종적으로 두 노모그램을 비교하고 효용성을 알아보았다. 마지막으로 ROC 곡선을 이용하여 노모그램이 적절한지 검증하였다.

Keywords

References

  1. Chung, S. M., Park, J. C., Moon, J. S., and Lee, J. Y. (2018). Novel nomogram for screening the risk of developing diabetes in a Korean population, Diabetes Research and Clinical Practice, 142, 286-293. https://doi.org/10.1016/j.diabres.2018.05.036
  2. Cook, N. R. (2008). Statistical evaluation of prognostic versus diagnostic models: beyond the ROC curve, Clinical Chemistry, 54, 17-23.
  3. Heo, M. H. and Lee, Y. G. (2008). Data Mining Modeling and Example, Hannarae, Seoul.
  4. Iasonos, A., Schrag, D., Raj, G. V., and Panageas, K. S. (2008). How to build and interpret a nomogram for cancer prognosis, Journal of Clinical Oncology, 26, 1364-1370. https://doi.org/10.1200/JCO.2007.12.9791
  5. Kim, Y. J., Jeon, J. Y., Han, S. J., Kim, H. J., Lee, K. W., and Kim, D. J. (2015). Effect of socio-economic status on the prevalence of diabetes, Yonsei Medical Journal, 56, 641-647. https://doi.org/10.3349/ymj.2015.56.3.641
  6. Korean Diabetes Association (2017). Korean diabetes fact sheet in Korea 2016. Publish: diabetes fact sheet in Korea, Available from: http://www.diabetes.or.kr/pro/news/admin.php?category=A&code=admin&number=1428&mode=view
  7. Lee, J. W., Park, M. R., and Yu, H. N. (2005). Statistical Method for Bioscience Research, Freedom academy, Seoul.
  8. Lee, K. M., Kim, W. J., and Yun, S. J. (2009). A clinical nomogram construction method using genetic algorithm and naive Bayesian technique, Journal of Korean Institute of Intelligent Systems, 19, 796-801. https://doi.org/10.5391/JKIIS.2009.19.6.796
  9. Lee, S. Y., Park, H. S., Kim, S. M., et al. (2006). Cut-off points of waist circumference for defining abdominal obesity in the Korean population, The Korean Journal of Obesity, 15, 1-9.
  10. Mozina, M., Demsar, J., Kattan, M., and Zupan, B. (2004). Nomogram for Visualization of Naive Bayesian Classifier, Knowledge Discovery in Databases: PKDD 2004, 4, 337-348.
  11. Park, J. C. and Lee, J. Y. (2017). Risk factors for type 2 diabetes among Korean adults in 2014, Quantitative Bio-Science, 36, 15-21. https://doi.org/10.22283/qbs.2017.36.1.15
  12. Park, J. C. (2018). Proposal of nomogram using logistic and Bayesian technique for type 2 diabetes (Master's thesis), Yeungnam University, Gyeongsan.
  13. Park, J. C. and Lee, J. Y. (2018). How to build nomogram for type 2 diabetes using a naive Bayesian classifier technique, Journal of Applied Statistics, 45, 2999-3011. https://doi.org/10.1080/02664763.2018.1450366
  14. Statistics Korea (2014). Causes of death statistics 2014. Policy News, Available from: http://kostat.go.kr/portal/korea/kor nw/3/index.board?bmode=read&aSeq=348541
  15. Yang, D. (2014). Build prognostic nomograms for risk assessment using SAS. In Proceedings of SAS Global Forum 2013, Available from: http://support.sas.com/resources/papers/proceedings13/264-2013.pdf