Clinical Nutrition Research

  • 제12권2호
  • /
  • Pages.138-153
  • /
  • 2023
  • /
  • 2287-3732(pISSN)
  • /
  • 2287-3740(eISSN)
한국임상영양학회 (The Korean Society of Clinical Nutrition)
권호 표지
DOI QR코드

DOI QR Code

Development of a Metabolic Syndrome Classification and Prediction Model for Koreans Using Deep Learning Technology: The Korea National Health and Nutrition Examination Survey (KNHANES) (2013-2018)

  • Hyerim Kim (Department of Food and Nutrition, Gyeongsang National University) ;
  • Ji Hye Heo (Department of Information & Statistics, Gyeongsang National University) ;
  • Dong Hoon Lim (Department of Information & Statistics, Research Institute of Natural Science (RINS), Gyeongsang National University) ;
  • Yoona Kim (Department of Food and Nutrition, Institute of Agriculture and Life Science, Gyeongsang National University)
  • 투고 : 2022.12.01
  • 심사 : 2023.03.27
  • 발행 : 2023.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

The prevalence of metabolic syndrome (MetS) and its cost are increasing due to lifestyle changes and aging. This study aimed to develop a deep neural network model for prediction and classification of MetS according to nutrient intake and other MetS-related factors. This study included 17,848 individuals aged 40-69 years from the Korea National Health and Nutrition Examination Survey (2013-2018). We set MetS (3-5 risk factors present) as the dependent variable and 52 MetS-related factors and nutrient intake variables as independent variables in a regression analysis. The analysis compared and analyzed model accuracy, precision and recall by conventional logistic regression, machine learning-based logistic regression and deep learning. The accuracy of train data was 81.2089, and the accuracy of test data was 81.1485 in a MetS classification and prediction model developed in this study. These accuracies were higher than those obtained by conventional logistic regression or machine learning-based logistic regression. Precision, recall, and F1-score also showed the high accuracy in the deep learning model. Blood alanine aminotransferase (β = 12.2035) level showed the highest regression coefficient followed by blood aspartate aminotransferase (β = 11.771) level, waist circumference (β = 10.8555), body mass index (β = 10.3842), and blood glycated hemoglobin (β = 10.1802) level. Fats (cholesterol [β = -2.0545] and saturated fatty acid [β = -2.0483]) showed high regression coefficients among nutrient intakes. The deep learning model for classification and prediction on MetS showed a higher accuracy than conventional logistic regression or machine learning-based logistic regression.

키워드

과제정보

The study was conducted according to the guidelines of the Declaration of Helsinki. Ethical review and approval were waived for this study, because KNHANES VI-VII was conducted by obtaining the approval of Research Ethics Review Committee of KCDC (2013-07CON-03-4C; 2013-12EXP-03-5C; 2018-01-03-P-A).

참고문헌

  1. Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, Fruchart JC, James WP, Loria CM, Smith SC Jr; International Diabetes Federation Task Force on Epidemiology and Prevention; Hational Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; International Association for the Study of Obesity. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120:1640-5.
  2. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365:1415-28.
  3. Samson SL, Garber AJ. Metabolic syndrome. Endocrinol Metab Clin North Am 2014;43:1-23.
  4. Ighbariya A, Weiss R. Insulin resistance, prediabetes, metabolic syndrome: what should every pediatrician know? J Clin Res Pediatr Endocrinol 2017;9:49-57.
  5. Reaven GM. Insulin resistance: the link between obesity and cardiovascular disease. Med Clin North Am 2011;95:875-92.
  6. Ferrannini E, Haffner SM, Mitchell BD, Stern MP. Hyperinsulinaemia: the key feature of a cardiovascular and metabolic syndrome. Diabetologia 1991;34:416-22.
  7. Mottillo S, Filion KB, Genest J, Joseph L, Pilote L, Poirier P, Rinfret S, Schiffrin EL, Eisenberg MJ. The metabolic syndrome and cardiovascular risk a systematic review and meta-analysis. J Am Coll Cardiol 2010;56:1113-32.
  8. Scuteri A, Najjar SS, Morrell CH, Lakatta EG; Cardiovascular Health Study. The metabolic syndrome in older individuals: prevalence and prediction of cardiovascular events: the Cardiovascular Health Study. Diabetes Care 2005;28:882-7.
  9. Wilson PW, D'Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation 2005;112:3066-72.
  10. Borch-Johnsen K. Metabolic syndrome in a global perspective. Significance for public health. Ugeskr Laeger 2006;168:3032-4.
  11. Ranasinghe P, Mathangasinghe Y, Jayawardena R, Hills AP, Misra A. Prevalence and trends of metabolic syndrome among adults in the Asia-Pacific region: a systematic review. BMC Public Health 2017;17:101.
  12. Hirode G, Wong RJ. Trends in the prevalence of metabolic syndrome in the United States, 2011-2016. JAMA 2020;323:2526-8.
  13. Lee SE, Han K, Kang YM, Kim SO, Cho YK, Ko KS, Park JY, Lee KU, Koh EH; Taskforce Team of Diabetes Fact Sheet of the Korean Diabetes Association. Trends in the prevalence of metabolic syndrome and its components in South Korea: Findings from the Korean National Health Insurance Service Database (2009-2013). PLoS One 2018;13:e0194490.
  14. Kim KY, Dong JY, Han SY, Lee KS. The effects of the metabolic syndrome on the total medical charge. Health Policy Manag 2017;27:47-55.
  15. Scholze J, Alegria E, Ferri C, Langham S, Stevens W, Jeffries D, Uhl-Hochgraeber K. Epidemiological and economic burden of metabolic syndrome and its consequences in patients with hypertension in Germany, Spain and Italy; a prevalence-based model. BMC Public Health 2010;10:529.
  16. Lee G, Choi HY. Factors associated with dietary control and physical activity in the management of metabolic syndrome in Korean menopausal women. Int J Environ Res Public Health 2020;17:6901.
  17. Sayon-Orea C, Razquin C, Bullo M, Corella D, Fito M, Romaguera D, Vioque J, Alonso-Gomez AM, Warnberg J, Martinez JA, Serra-Majem L, Estruch R, Tinahones FJ, Lapetra J, Pinto X, Tur JA, LopezMiranda J, Bueno-Cavanillas A, Delgado-Rodriguez M, Matia-Martin P, Daimiel L, Sanchez VM, Vidal J, Vazquez C, Ros E, Ruiz-Canela M, Sorli JV, Castaner O, Fiol M, Navarrete-Munoz EM, Aros F, GomezGracia E, Zulet MA, Sanchez-Villegas A, Casas R, Bernal-Lopez R, Santos-Lozano JM, Corbella E, Bouzas C, Garcia-Arellano A, Basora J, Asensio EM, Schroder H, Monino M, Garcia de la Hera M, Tojal-Sierra L, Toledo E, Diaz-Lopez A, Goday A, Salas-Salvado J, Martinez-Gonzalez MA. Effect of a nutritional and behavioral intervention on energy-reduced Mediterranean diet adherence among patients with metabolic syndrome: interim analysis of the PREDIMED-Plus randomized clinical trial. JAMA 2019;322:1486-99.
  18. Riccardi G, Rivellese AA. Dietary treatment of the metabolic syndrome--the optimal diet. Br J Nutr 2000;83 Suppl 1:S143-8.
  19. Agodi A, Maugeri A, Kunzova S, Sochor O, Bauerova H, Kiacova N, Barchitta M, Vinciguerra M. Association of dietary patterns with metabolic syndrome: results from the Kardiovize Brno 2030 study. Nutrients 2018;10:898.
  20. Cho YA, Kim J, Cho ER, Shin A. Dietary patterns and the prevalence of metabolic syndrome in Korean women. Nutr Metab Cardiovasc Dis 2011;21:893-900.
  21. Soltani S, Kolahdouz Mohammadi R, Shab-Bidar S, Vafa M, Salehi-Abargouei A. Sodium status and the metabolic syndrome: a systematic review and meta-analysis of observational studies. Crit Rev Food Sci Nutr 2019;59:196-206.
  22. Castro-Barquero S, Ruiz-Leon AM, Sierra-Perez M, Estruch R, Casas R. Dietary strategies for metabolic syndrome: a comprehensive review. Nutrients 2020;12:2983.
  23. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015;521:436-44.
  24. Charniak E. Introduction to deep learning. Cambridge (MA): The MIT Press; 2019.
  25. Chollet F. Deep learning with python. Shelter Island (NY): Manning Publications Company; 2017.
  26. Choe EK, Rhee H, Lee S, Shin E, Oh SW, Lee JE, Choi SH. Metabolic syndrome prediction using machine learning models with genetic and clinical information from a nonobese healthy population. Genomics Inform 2018;16:e31.
  27. Kim J, Mun S, Lee S, Jeong K, Baek Y. Prediction of metabolic and pre-metabolic syndromes using machine learning models with anthropometric, lifestyle, and biochemical factors from a middle-aged population in Korea. BMC Public Health 2022;22:664.
  28. Kweon S, Kim Y, Jang MJ, Kim Y, Kim K, Choi S, Chun C, Khang YH, Oh K. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol 2014;43:69-77.
  29. Korea Disease Control and Prevention Agency. Survey contents South Korea [Internet]. Available from https://knhanes.cdc.go.kr/knhanes/sub02/sub02_03.do#s8_04 [cited 2022 January 3]. 2022.
  30. Korea Disease Control and Prevention Agency. Findings from Korea National Health and Nutrition Examination Survey [Internet]. Available from https://knhanes.cdc.go.kr/knhanes/sub01/sub01_05.do#s5_02 [cited 2021 June 1]. 2018.
  31. Kim J, Kim Y, Ahn YO, Paik HY, Ahn Y, Tokudome Y, Hamajima N, Inoue M, Tajima K. Development of a food frequency questionnaire in Koreans. Asia Pac J Clin Nutr 2003;12:243-50.
  32. Kim DW, Song S, Lee JE, Oh K, Shim J, Kweon S, Paik HY, Joung H. Reproducibility and validity of an FFQ developed for the Korea National Health and Nutrition Examination Survey (KNHANES). Public Health Nutr 2015;18:1369-77.
  33. Nguyen N, Cripps A. Predicting housing value: a comparison of multiple regression analysis and artificial neural networks. J Real Estate Res 2001;22:313-36.
  34. Lee SH, Tao S, Kim HS. The prevalence of metabolic syndrome and its related risk complications among Koreans. Nutrients 2019;11:1755.
  35. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith SC Jr, Spertus JA, Costa F; American Heart Association; National Heart, Lung, and Blood Institute. Diagnosis and management of the metabolic syndrome: an American Heart Association/ National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005;112:2735-52.
  36. Enkhmaa B, Shiwaku K, Anuurad E, Nogi A, Kitajima K, Yamasaki M, Oyunsuren T, Yamane Y. Prevalence of the metabolic syndrome using the Third Report of the National Cholesterol Educational Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (ATP III) and the modified ATP III definitions for Japanese and Mongolians. Clin Chim Acta 2005;352:105-13.
  37. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998;15:539-53.
  38. Korean Society for the Study of Obesity. Guideline for the management of obesity 2018. Seoul: Korean Society for the Study of Obesity; 2018.
  39. Qasim A, Turcotte M, de Souza RJ, Samaan MC, Champredon D, Dushoff J, Speakman JR, Meyre D. On the origin of obesity: identifying the biological, environmental and cultural drivers of genetic risk among human populations. Obes Rev 2018;19:121-49.
  40. Yu CS, Lin YJ, Lin CH, Wang ST, Lin SY, Lin SH, Wu JL, Chang SS. Predicting metabolic syndrome with machine learning models using a decision tree algorithm: retrospective cohort study. JMIR Med Inform 2020;8:e17110.
  41. Worachartcheewan A, Shoombuatong W, Pidetcha P, Nopnithipat W, Prachayasittikul V, Nantasenamat C. Predicting metabolic syndrome using the random forest method. Sci World J 2015;2015:581501.
  42. Park JE, Mun S, Lee S. Metabolic syndrome prediction models using machine learning and Sasang constitution type. Evid Based Complement Alternat Med 2021;2021:8315047.
  43. Hanley AJ, Williams K, Festa A, Wagenknecht LE, D'Agostino RB Jr, Haffner SM. Liver markers and development of the metabolic syndrome: the insulin resistance atherosclerosis study. Diabetes 2005;54:3140-7.
  44. Lonardo A, Ballestri S, Marchesini G, Angulo P, Loria P. Nonalcoholic fatty liver disease: a precursor of the metabolic syndrome. Dig Liver Dis 2015;47:181-90.
  45. Bozzetto L, Prinster A, Annuzzi G, Costagliola L, Mangione A, Vitelli A, Mazzarella R, Longobardo M, Mancini M, Vigorito C, Riccardi G, Rivellese AA. Liver fat is reduced by an isoenergetic MUFA diet in a controlled randomized study in type 2 diabetic patients. Diabetes Care 2012;35:1429-35.
  46. Phelan S, Wadden TA, Berkowitz RI, Sarwer DB, Womble LG, Cato RK, Rothman R. Impact of weight loss on the metabolic syndrome. Int J Obes 2007;31:1442-8.
  47. Finucane OM, Lyons CL, Murphy AM, Reynolds CM, Klinger R, Healy NP, Cooke AA, Coll RC, McAllan L, Nilaweera KN, O'Reilly ME, Tierney AC, Morine MJ, Alcala-Diaz JF, Lopez-Miranda J, O'Connor DP, O'Neill LA, McGillicuddy FC, Roche HM. Monounsaturated fatty acid-enriched high-fat diets impede adipose NLRP3 inflammasome-mediated IL-1β secretion and insulin resistance despite obesity. Diabetes 2015;64:2116-28.
  48. Lee TC, Ivester P, Hester AG, Sergeant S, Case LD, Morgan T, Kouba EO, Chilton FH. The impact of polyunsaturated fatty acid-based dietary supplements on disease biomarkers in a metabolic syndrome/diabetes population. Lipids Health Dis 2014;13:196.
  49. Julibert A, Bibiloni MD, Tur JA. Dietary fat intake and metabolic syndrome in adults: a systematic review. Nutr Metab Cardiovasc Dis 2019;29:887-905.
  50. Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev 2020;5:CD011737.
  51. Sellem L, Flourakis M, Jackson KG, Joris PJ, Lumley J, Lohner S, Mensink RP, Soedamah-Muthu SS, Lovegrove JA. Impact of replacement of individual dietary sfas on circulating lipids and other biomarkers of cardiometabolic health: a systematic review and meta-analysis of randomized controlled trials in humans. Adv Nutr 2022;13:1200-25.
  52. Siri-Tarino PW, Sun Q, Hu FB, Krauss RM. Saturated fatty acids and risk of coronary heart disease: modulation by replacement nutrients. Curr Atheroscler Rep 2010;12:384-90.
  53. Mozaffarian D, Micha R, Wallace S. Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat: a systematic review and meta-analysis of randomized controlled trials. PLoS Med 2010;7:e1000252.
  54. Jang H, Park K. Omega-3 and omega-6 polyunsaturated fatty acids and metabolic syndrome: a systematic review and meta-analysis. Clin Nutr 2020;39:765-73.
  55. Kothawade PB, Thomas AB, Chitlange SS. Novel niacin receptor agonists: a promising strategy for the treatment of dyslipidemia. Mini Rev Med Chem 2021;21:2481-96.
  56. Boden WE, Sidhu MS, Toth PP. The therapeutic role of niacin in dyslipidemia management. J Cardiovasc Pharmacol Ther 2014;19:141-58.
  57. Kim SW, Kim HJ, Min K, Lee H, Lee SH, Kim S, Kim JS, Oh B. The relationship between smoking cigarettes and metabolic syndrome: a cross-sectional study with non-single residents of Seoul under 40 years old. PLoS One 2021;16:e0256257.
  58. Kim T, Kang J. Association between serum retinol and α-tocopherol levels and metabolic syndrome in Korean general population: analysis of population-based nationally representative data. Nutrients 2020;12:1689.