Journal of the Korea Academia-Industrial cooperation Society (한국산학기술학회논문지)

The Korea Academia-Industrial cooperation Society (한국산학기술학회)
권호 표지
DOI QR코드

DOI QR Code

A study on the development of severity-adjusted mortality prediction model for discharged patient with acute stroke using machine learning

머신러닝을 이용한 급성 뇌졸중 퇴원 환자의 중증도 보정 사망 예측 모형 개발에 관한 연구

  • 백설경 (아주대학교병원) ;
  • 박종호 (계명대학교 동산의료원) ;
  • 강성홍 (인제대학교 보건행정학과) ;
  • 박혜진 (대구가톨릭대학교 국제의료경영학과)
  • Received : 2018.08.13
  • Accepted : 2018.11.02
  • Published : 2018.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study was to develop a severity-adjustment model for predicting mortality in acute stroke patients using machine learning. Using the Korean National Hospital Discharge In-depth Injury Survey from 2006 to 2015, the study population with disease code I60-I63 (KCD 7) were extracted for further analysis. Three tools were used for the severity-adjustment of comorbidity: the Charlson Comorbidity Index (CCI), the Elixhauser comorbidity index (ECI), and the Clinical Classification Software (CCS). The severity-adjustment models for mortality prediction in patients with acute stroke were developed using logistic regression, decision tree, neural network, and support vector machine methods. The most common comorbid disease in stroke patients were hypertension, uncomplicated (43.8%) in the ECI, and essential hypertension (43.9%) in the CCS. Among the CCI, ECI, and CCS, CCS had the highest AUC value. CCS was confirmed as the best severity correction tool. In addition, the AUC values for variables of CCS including main diagnosis, gender, age, hospitalization route, and existence of surgery were 0.808 for the logistic regression analysis, 0.785 for the decision tree, 0.809 for the neural network and 0.830 for the support vector machine. Therefore, the best predictive power was achieved by the support vector machine technique. The results of this study can be used in the establishment of health policy in the future.

본 연구는 머신러닝을 활용하여 급성 뇌졸중 퇴원 환자의 중증도 보정 사망 예측 모형 개발을 목적으로 시행하였다. 전국 단위의 퇴원손상심층조사 2006~2015년 자료 중 한국표준질병사인분류(Korean standard classification of disease-KCD 7)에 따라 뇌졸중 코드 I60-I63에 해당하는 대상자를 추출하여 분석하였다. 동반질환 중증도 보정 도구로는 Charlson comorbidity index(CCI), Elixhauser comorbidity index(ECI), Clinical classification software(CCS)의 3가지 도구를 사용하였고 중증도 보정 모형 예측 개발은 로지스틱회귀분석, 의사결정나무, 신경망, 서포트 벡터 머신 기법을 활용하여 비교해 보았다. 뇌졸중 환자의 동반질환으로는 ECI에서는 합병증을 동반하지 않은 고혈압(hypertension, uncomplicated)이 43.8%로, CCS에서는 본태성고혈압(essential hypertension)이 43.9%로 다른 질환에 비해 가장 월등하게 높은 것으로 나타났다. 동반질환 중중도 보정 도구를 비교해 본 결과 CCI, ECI, CCS 중 CCS가 가장 높은 AUC값으로 분석되어 가장 우수한 중증도 보정 도구인 것으로 확인되었다. 또한 CCS, 주진단, 성, 연령, 입원경로, 수술유무 변수를 포함한 중증도 보정 모형 개발 AUC값은 로지스틱 회귀분석의 경우 0.808, 의사결정나무 0.785, 신경망 0.809, 서포트 벡터 머신 0.830로 분석되어 가장 우수한 예측력을 보인 것은 서포트 벡터머신 기법인 것으로 최종 확인되었고 이러한 결과는 추후 보건의료정책 수립에 활용될 수 있을 것이다.

Keywords

SHGSCZ_2018_v19n11_126_f0001.png 이미지

Fig. 1. Logistic regression model development using KNIME

SHGSCZ_2018_v19n11_126_f0002.png 이미지

Fig. 2. Severity-adjusted mortality rate model for acute stroke patients using decision tree

SHGSCZ_2018_v19n11_126_f0003.png 이미지

Fig. 3. Neural network model development using KNIME

SHGSCZ_2018_v19n11_126_f0004.png 이미지

Fig. 4. Support vector machine model development using KNIME

Table 1. Definition of variables

SHGSCZ_2018_v19n11_126_t0001.png 이미지

Table 2. General characteristics of acute stroke inpatients

SHGSCZ_2018_v19n11_126_t0002.png 이미지

Table 3. Distribution of principal diagnosis

SHGSCZ_2018_v19n11_126_t0003.png 이미지

Table 4. Distribution of Charlson comorbidity index

SHGSCZ_2018_v19n11_126_t0004.png 이미지

Table 5. Distribution of comorbidity disease by Elixhauser comorbidity index

SHGSCZ_2018_v19n11_126_t0005.png 이미지

Table 6. Distribution of comorbidity disease by clinical classification software category

SHGSCZ_2018_v19n11_126_t0006.png 이미지

Table 7. Logistic regression model assessment using AUC

SHGSCZ_2018_v19n11_126_t0007.png 이미지

Table 8. Severity-adjusted mortality rate model for acute stroke patients using logistic regression

SHGSCZ_2018_v19n11_126_t0008.png 이미지

Table 9. Decision Tree model assessment using AUC

SHGSCZ_2018_v19n11_126_t0009.png 이미지

Table 10. Neural network model assessment using AUC

SHGSCZ_2018_v19n11_126_t0010.png 이미지

Table 11. Support vector machine model assessment using AUC

SHGSCZ_2018_v19n11_126_t0011.png 이미지

References

  1. J. H. Im, K. S. Lee, K. Y. Kim, N. S. Hong, S. W. Lee, H. J. Bae, "Follow-up study on mortality in Korean stroke patients", J Korean Med Assoc, Vol. 54 No. 11 pp. 1199-1208, 2011. DOI: https://dx.doi.org/10.5124/jkma.2011.54.11.1199
  2. K. J. Kim, M. Heo, I. A. Chun, H. J. Jun, J. S. Lee, H, Jegal, "The re-lationship between stroke and quality of life in Korean adults: based on the 2010 Korean community health survey". Journal of Physical Therapy Science, Vol. 27, No. 1, pp. 309-312, 2015. https://doi.org/10.1589/jpts.27.309
  3. E. H. Jeong, O. T, M. S, B. R. Kim, J. Lee, "Relationship between Comorbid Cognitive Impairment and Functional Outcomes in Stroke Patients with Spatial Neglect", Brain & Neuro Rehabilitation, Vol. 9, No. 1, 2016.
  4. H. Rubin, P. Pronovost, G. Diette, "The advantages and disadvantages of process-based measure of health care quality", International Journal for Quality in Health Care, Vol. 13, pp. 469-474, 2001. DOI: https://dx.doi.org/10.1093/intqhc/13.6.469
  5. J. Y. Yoon, J. Y. Lee, "Development of Outcome Indicators of Urinary Incontinence for Quality Evaluation in Long Term Care Hospitals", J Korean Acad Nurs. Vol. 40, No. 1, pp. 110-118, 2010. DOI: https://dx.doi.org/10.4040/jkan.2010.40.1.110
  6. J. H. Park, Y. M. Kim, S. S. Kim, W. J. Kim, S. H. K, "Comparison of Hospital Standardized Mortality Ratio Using National Hospital Discharge Injury Data", Journal of the Korea Academia-Industrial cooperation Society, Vol. 13, No. 4 pp. 1739-1750, 2012. DOI: https://dx.doi.org/10.5762/KAIS.2012.13.4.1739
  7. A. Elixhauser, C. Steiner, D. Harris, R. Coffey, "Comorbidity measures for use with administrative data", Med Care, Vol. 36, pp. 8-27, 1998. DOI: https://dx.doi.org/10.1097/00005650-199801000-00004
  8. C. Y. Wang, Y. S. Lin, C. Tzao, H. C. Lee, M. H. Huang, W. H. Hsu, H. S. Hsu, "Comparison of Charlson comorbidity index and Kaplan-Feinstein index in patients with stage I lung cancer after surgical resection", European Journal of Cardio-Thoracic Surgery, Vol.32, No.1 pp. 877-881, December 2007. DOI: https://dx.doi.org/10.1016/j.ejcts.2007.09.008
  9. M. Charlson, P. Pompei, K. Ales, C. MacKenzie, "A new method of classifying prognostic comorbidity in longitudinal studies: development and validation", J Chronic Dis, Vol. 40, pp. 373-383, 1987. DOI: https://dx.doi.org/10.1016/0021-9681(87)90171-8
  10. S. H. Kang, "Analysis of LOS variance-The results of Korean National Hospital Discharge In-depth Injury Survey 2004-2006", Public health weekly report, KCDC, 2, pp. 587-866, 2009.
  11. Y. M. Kim, "A study on analysis of factors on in-hospital mortality for community-acquired pneumonia", Journal of the Korean Data & Information Science Society, Vol. 22, No. 3, pp. 389-400, 2011.
  12. M. H. Nam, S. S. Kim, I. S. Park, S. H. Kang, W. J. Kim, S. H. Choi, H. K. Jo, Y. T. Kim, S. O. Hong, "A Study on Utilization of non-residential areal in Operation patient", Journal of the Korea Academia-Industrial cooperation Society, Vol. 11, No. 6, pp. 2078-2087, 2010. DOI: https://dx.doi.org/10.5762/KAIS.2010.11.6.2078
  13. Australia, National Health Data Committee, National Health Data Dictionary, Version 12, AIHW, 2003.
  14. http://mchp-appserv.cpe.umanitoba.ca/viewConcept.php? printer=Y&conceptID=1098
  15. J. H. Lee, J. K. Choi, S. N. Jeong, S. H. Choi. "Charlson comorbidity index as a predictor of periodontal disease in elderly participants" J Periodontal Implant Sci, Vol. 48, No. 2, pp. 92-102, 2018. https://doi.org/10.5051/jpis.2018.48.2.92
  16. http://mchp-appserv.cpe.umanitoba.ca/viewConcept.php?printer=Y&conceptID=1436
  17. https://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp
  18. Y. H. Choi, Y. J. Kim, S. O. Hong, "Development of Severity-adjusted Length of Stay in Ischemic Heart Disease", Journal of the Korean Data Analysis Society, Vol. 17, No. 1, pp. 407-421, 2015.
  19. K. H. Kim, J. H. Im, "Mortality of Stroke Patients Based on Charlson comorbidity index", Journal of the Korea Contents Association, Vol. 16, No.3, pp. 22-32, 2016. DOI: https://dx.doi.org/10.5392/JKCA.2016.16.03.022
  20. J. George, S. Douglas, P. Wagner, F. Alfred, "Comparison of the performance of two comorbidity measures, with and without information from prior hospitalizations.", Medical Care, Vol. 39, No. 7, pp. 727-739, 2001. https://doi.org/10.1097/00005650-200107000-00009
  21. D. A. Southern, H. Quan, W. A. Ghali, "Comparison of the Elixhauser and Charlson/Deyo Methods of Comorbidity Measurement in Administrative Data", Medical Care, Vol. 42, No. 4, pp. 355-360, 2004. DOI: https://dx.doi.org/10.1097/01.mlr.0000118861.56848.ee
  22. P. R. Hachesu, M. Ahmadi, S. Alizadeh, F. Sadoughi, "Use of data mining techniques to determine and predict length of stay of cardiac patients", Healthc Inform Res Vol. 19, No. 2, pp. 121-9, 2013. DOI: https://dx.doi.org/10.4258/hir.2013.19.2.121
  23. L. Tapak, H. Mahjub, O. Hamidi, J. Poorolajal, "Real-Data Comparison of Data Mining Methods in Prediction of Diabetes in Iran", Healthc Inform Res. Vol. 19, No. 3, pp. 177-185, 2013. DOI: https://dx.doi.org/10.4258/hir.2013.19.3.177
  24. C. Lehmann, T. Koenig, V. Jelic, L. Prichep, "Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG)", J Neurosci Methods, Vol. 161, No. 2, pp. 342-350, 2007. DOI: https://dx.doi.org/10.1016/j.jneumeth.2006.10.023