The Journal of the Convergence on Culture Technology (문화기술의 융합)

The International Promotion Agency of Culture Technology (국제문화기술진흥원)
권호 표지
DOI QR코드

DOI QR Code

Prediction model of hypercholesterolemia using body fat mass based on machine learning

머신러닝 기반 체지방 측정정보를 이용한 고콜레스테롤혈증 예측모델

  • Lee, Bum Ju (Dept. of Future Medicine Division, Korea Institute of Oriental Medicine)
  • Received : 2019.09.15
  • Accepted : 2019.10.20
  • Published : 2019.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of the present study is to develop a model for predicting hypercholesterolemia using an integrated set of body fat mass variables based on machine learning techniques, beyond the study of the association between body fat mass and hypercholesterolemia. For this study, a total of six models were created using two variable subset selection methods and machine learning algorithms based on the Korea National Health and Nutrition Examination Survey (KNHANES) data. Among the various body fat mass variables, we found that trunk fat mass was the best variable for predicting hypercholesterolemia. Furthermore, we obtained the area under the receiver operating characteristic curve value of 0.739 and the Matthews correlation coefficient value of 0.36 in the model using the correlation-based feature subset selection and naive Bayes algorithm. Our findings are expected to be used as important information in the field of disease prediction in large-scale screening and public health research.

본 연구의 목적은 기존의 body fat mass 변수와 고콜레스테롤혈증의 연관성연구를 벗어나, 머신러닝기법을 기반으로 body fat mass 변수들의 조합을 이용하여 고콜레스테롤혈증 예측 모델을 개발하는 것이다. 이러한 연구를 위하여 국민건강영양조사 데이터를 기반으로 두 가지 variable selection 메소드와 머신러닝 알고리즘을 이용하여 총 6개의 모델을 생성하였고 질병 예측력을 비교분석하였다. 여러 body fat mass 관련 변수들 중에서 몸통지방량 변수가 고콜레스테롤혈증 예측력이 가장 우수한 변수인 것을 밝혀내었고, 머신러닝 기반 예측모델들 중에서 correlation-based feature subset selection 기반 naive Bayes 알고리즘을 이용한 모델이 0.739의 the area under the receiver operating characteristic curve 값과 0.36의 Matthews correlation coefficient 값을 얻었다. 이러한 연구의 결과는 향후 국내외 대규모 스크리닝 및 대중보건 연구에서 질병예측분야의 중요정보로 활용될 것으로 예상한다.

Keywords

References

  1. Berbee JF, Boon MR, Khedoe PP, Bartelt A, Schlein C, Worthmann A, Kooijman S, Hoeke G, Mol IM, John C, Jung C, Vazirpanah N, Brouwers LP, Gordts PL, Esko JD, Hiemstra PS, Havekes LM, Scheja L, Heeren J, Rensen PC. Brown fat activation reduces hypercholesterolaemia and protects from atherosclerosis development. Nat Commun. 2015;6:6356. doi: 10.1038/ncomms7356.
  2. Trentman TL, Avey SG, Ramakrishna H. Current and emerging treatments for hypercholesterolemia: A focus on statins and proprotein convertase subtilisin/kexin Type 9 inhibitors for perioperative clinicians. J Anaesthesiol Clin Pharmacol. 2016;32(4):440-445. doi: 10.4103/0970-9185.194773.
  3. Knowles JW, Rader DJ, Khoury MJ. Cascade Screening for Familial Hypercholesterolemia and the Use of Genetic Testing. JAMA. 2017;318(4):381-382. doi: 10.1001/jama.2017.8543.
  4. Lee BJ, Ku B, A comparison of trunk circumference and width indices for hypertension and type 2 diabetes in a large-scale screening: a retrospective cross-sectional study. Sci Rep. 2018;8:13284(1-10). doi: 10.1038/s41598-018-31624-x
  5. Lee BJ, Kim JY. Identification of Type 2 Diabetes Risk Factors Using Phenotypes Consisting of Anthropometry and Triglycerides based on Machine Learning. IEEE J Biomed Health Inform. 2016;20(1):39-46. doi: 10.1109/JBHI.2015.2396520.
  6. Lee BJ, Kim JY. Identification of the Best Anthropometric Predictors of Serum High- and Low-Density Lipoproteins Using Machine Learning. IEEE J Biomed Health Inform. 2015;19(5):1747-1756. doi: 10.1109/JBHI.2014.2350014.
  7. Lee BJ, Kim JY. Indicators of hypertriglyceridemia from anthropometric measures based on data mining. Comput Biol Med. 2015;57:201-211. doi: 10.1016/j.compbiomed.2014.12.005.
  8. Lee BJ, Kim JY. A comparison of the predictive power of anthropometric indices for hypertension and hypotension risk. PLoS One 2014;9(1):e84897. doi: 10.1371/journal.pone.0084897.
  9. Lee BJ, Ku B, Nam J, Pham DD, Kim JY. Prediction of fasting plasma glucose status using anthropometric measures for diagnosing type 2 diabetes. IEEE J Biomed Health Inform. 2014;18(2):555-561. doi: 10.1109/JBHI.2013.2264509.
  10. Lee BJ, Kim JY. Identification of Hemoglobin Levels Based on Anthropometric Indices in Elderly Koreans. PLoS One 2016;11(11):e0165622. doi: 10.1371/journal.pone.0165622.
  11. Chi JH, Shin MS, Lee BJ. Association of type 2 diabetes with anthropometrics, bone mineral density, and body composition in a large-scale screening study of Korean adults. PLoS One. 2019;14(7):e0220077. doi:10.1371/journal.pone.0220077.
  12. Ahn E, Kim E. A study on the eating behaviors and food intake of diabetic patients in Daegu.Gyeongbuk area. The Journal of the Convergence on Culture Technology. 2019;5(3):229-239 doi: http://dx.doi.org/10.17703/JCCT.2019.5.229.
  13. Vasan SK, Osmond C, Canoy D, Christodoulides C, Neville MJ, Di Gravio C, Fall CHD, Karpe F. Comparison of regional fat measurements by dual-energy X-ray absorptiometry and conventional anthropometry and their association with markers of diabetes and cardiovascular disease risk. Int J Obes (Lond). 2018;42(4):850-857. doi: 10.1038/ijo.2017.289.
  14. Gastaldelli A. Abdominal fat: does it predict the development of type 2 diabetes? Am J Clin Nutr. 2008;87(5):1118-1119. doi: 10.1093/ajcn/87.5.1118
  15. Ohlson LO, Larsson B, Svardsudd K, Welin L, Eriksson H, Wilhelmsen L, et al. The influence of body fat distribution on the incidence of diabetes mellitus: 13.5 years of follow-up of the participants in the study of men born in 1913. Diabetes. 1985;34(10):1055-1058. doi: 10.2337/diab.34.10.1055
  16. Carey VJ, Walters EE, Colditz GA, Solomon CG, Willet WC, Rosner BA, et al. Body fat distribution and risk of non-insulin-dependent diabetes mellitus in women: the Nurses' Health Study. Am J Epidemiol. 1997;145(7):614-619. doi: 10.1093/oxfordjournals.aje.a009158
  17. Ortega FB, Sui X, Lavie CJ, Blair SN. Body Mass Index, the Most Widely Used but also Widely Criticized Index: Would a Gold-Standard Measure of Total Body Fat be a Better Predictor of Cardiovascular Disease Mortality? Mayo Clin Proc. 2016;91(4):443-455. doi: 10.1016/j.mayocp.2016.01.008
  18. Sookyung Hyun, Susan Moffatt-Bruce, Cheryl Newton, Brenda Hixon, Pacharmon Kaewprag. Tree-based Approach to Predict Hospital Acquired Pressure Injury. International Journal of Advanced Culture Technology. 2019;7(1):8-13 doi: 10.17703/IJACT.2019.7.1.8.
  19. Gishti O, Gaillard R, Durmus B, Abrahamse M, van der Beek EM, Hofman A, Franco OH, de Jonge LL, Jaddoe VW. BMI, total and abdominal fat distribution, and cardiovascular risk factors in school-age children. Pediatr Res. 2015;77(5):710-718. doi: 10.1038/pr.2015.29.
  20. Muls E, Kolanowski J, Scheen A, Van Gaal L; ObelHyx Study Group. The effects of orlistat on weight and on serum lipids in obese patients with hypercholesterolemia: a randomized, double-blind, placebo-controlled, multicentre study. Int J Obes Relat Metab Disord. 2001;25(11):1713-1721. https://doi.org/10.1038/sj.ijo.0801814
  21. Hecker KD, Kris-Etherton PM, Zhao G, Coval S, Jeor SS. Impact of body weight and weight loss on cardiovascular risk factors. Curr Atheroscler Rep. 1999;1:236-242. https://doi.org/10.1007/s11883-999-0038-2
  22. Wiklund P, Toss F, Weinehall L, Hallmans G, Franks PW, Nordstrom A, Nordstrom P. Abdominal and gynoid fat mass are associated with cardiovascular risk factors in men and women. J Clin Endocrinol Metab. 2008;93(11):4360-4366. doi: 10.1210/jc.2008-0804.
  23. The Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV-3), 2009, Korea Centers for Disease Control and Prevention.
  24. Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH. The WEKA data mining software: an update. SIGKDD Explor. 2009;1(1):10-18.
  25. Hall M, Holmes G. Benchmarking attribute selection techniques for discrete data class data mining. IEEE Trans Knowl Data Eng. 2003;15(6):1437-1447. https://doi.org/10.1109/TKDE.2003.1245283
  26. Kohavi R, John GH. Wrappers for feature subset selection. Artif Intell. 1997;97(1):273-324. https://doi.org/10.1016/S0004-3702(97)00043-X
  27. Bosy-Westphal A, Geisler C, Onur S, Korth O, Selberg O, Schrezenmeir J, Muller MJ. Value of body fat mass vs anthropometric obesity indices in the assessment of metabolic risk factors. Int J Obes (Lond). 2006;30(3):475-483. https://doi.org/10.1038/sj.ijo.0803144
  28. Lara-Esqueda A, Aguilar-Salinas CA, Velazquez-Monroy O, Gomez-Perez FJ, Rosas-Peralta M, Mehta R, Tapia-Conyer R. The body mass index is a less-sensitive tool for detecting cases with obesity-associated co-morbidities in short stature subjects. Int J Obes Relat Metab Disord. 2004;28(11):1443-1450. https://doi.org/10.1038/sj.ijo.0802705
  29. Gibby JT, Njeru DK, Cvetko ST, Merrill RM, Bikman BT, Gibby WA. Volumetric analysis of central body fat accurately predicts incidence of diabetes and hypertension in adults. BMC Obes. 2015;2:10. doi: 10.1186/s40608-015-0039-3.
  30. Gangwisch JE, Malaspina D, Babiss LA, Opler MG, Posner K, Shen S, Turner JB, Zammit GK, Ginsberg HN. Short sleep duration as a risk factor for hypercholesterolemia: analyses of the National Longitudinal Study of Adolescent Health. Sleep. 2010;33(7):956-961. https://doi.org/10.1093/sleep/33.7.956
  31. Shabnam AA, Homa K, Reza MT, Bagher L, Hossein FM, Hamidreza A. Cut-off points of waist circumference and body mass index for detecting diabetes, hypercholesterolemia and hypertension according to National Non-Communicable Disease Risk Factors Surveillance in Iran. Arch Med Sci. 2012;8(4):614-621. doi: 10.5114/aoms.2012.30284.