Machine Learning for the Prediction of New-Onset Diabetes Mellitus during 5-Year Follow-up in Non-Diabetic Patients with Cardiovascular Risks
1
조회
0
다운로드
6
인용
0
참고문헌
키워드
초록
Purpose: Many studies have proposed predictive models for type 2 diabetes mellitus (T2DM). However, these predictive modelshave several limitations, such as user convenience and reproducibility. The purpose of this study was to develop a T2DM predictivemodel using electronic medical records (EMRs) and machine learning and to compare the performance of this model withtraditional statistical methods. Materials and Methods: In this study, a total of available 8454 patients who had no history of diabetes and were treated at the cardiovascularcenter of Korea University Guro Hospital were enrolled. All subjects completed 5 years of follow up. The prevalence ofT2DM during follow up was 4.78% (404/8454). A total of 28 variables were extracted from the EMRs. In order to verify the crossvalidationtest according to the prediction model, logistic regression (LR), linear discriminant analysis (LDA), quadratic discriminantanalysis (QDA), and K-nearest neighbor (KNN) algorithm models were generated. The LR model was considered as the existingstatistical analysis method. Results: All predictive models maintained a change within the standard deviation of area under the curve (AUC) <0.01 in theanalysis after a 10-fold cross-validation test. Among all predictive models, the LR learning model showed the highest prediction performance,with an AUC of 0.78. However, compared to the LR model, the LDA, QDA, and KNN models did not show a statisticallysignificant difference. Conclusion: We successfully developed and verified a T2DM prediction system using machine learning and an EMR database,and it predicted the 5-year occurrence of T2DM similarly to with a traditional prediction model. In further study, it is necessary toapply and verify the prediction model through clinical research.
Abstract
Purpose: Many studies have proposed predictive models for type 2 diabetes mellitus (T2DM). However, these predictive modelshave several limitations, such as user convenience and reproducibility. The purpose of this study was to develop a T2DM predictivemodel using electronic medical records (EMRs) and machine learning and to compare the performance of this model withtraditional statistical methods. Materials and Methods: In this study, a total of available 8454 patients who had no history of diabetes and were treated at the cardiovascularcenter of Korea University Guro Hospital were enrolled. All subjects completed 5 years of follow up. The prevalence ofT2DM during follow up was 4.78% (404/8454). A total of 28 variables were extracted from the EMRs. In order to verify the crossvalidationtest according to the prediction model, logistic regression (LR), linear discriminant analysis (LDA), quadratic discriminantanalysis (QDA), and K-nearest neighbor (KNN) algorithm models were generated. The LR model was considered as the existingstatistical analysis method. Results: All predictive models maintained a change within the standard deviation of area under the curve (AUC) <0.01 in theanalysis after a 10-fold cross-validation test. Among all predictive models, the LR learning model showed the highest prediction performance,with an AUC of 0.78. However, compared to the LR model, the LDA, QDA, and KNN models did not show a statisticallysignificant difference. Conclusion: We successfully developed and verified a T2DM prediction system using machine learning and an EMR database,and it predicted the 5-year occurrence of T2DM similarly to with a traditional prediction model. In further study, it is necessary toapply and verify the prediction model through clinical research.
참고문헌 (28)
- American Diabetes Association, "Diagnosis and classification of diabetes mellitus", Diabetes Care, 36(Suppl 1), pp.S67~74 (2013)
- Wexler DJ, "Correlates of health-related quality of life in type 2diabetes", Diabetologia, 49, pp.1489~1497 (2006)
- Laakso M, "Hyperglycemia and cardiovascular disease in type 2diabetes", Diabetes, 48, pp.937~942 (1999)
- Romero SP, "Impact of new-onset diabetes mellitus and glycemic control on the prognosis of heart failure patients : a propensity-matched study in the community", Int J Cardiol, 167, pp.1206~1216 (2013)
- Twito O, "New-onset diabetes in elderly subjects : association between HbA1c levels, mortality, and coronary revascularization", Diabetes Care, 36, pp.3425~3429 (2013)
- Tuomilehto J, "Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance", N Engl J Med, 344, pp.1343~1350 (2001)
- Knowler WC, "Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin", N Engl J Med, 346, pp.393~403 (2002)
- Park JY, "Impact of low dose atorvastatin on development of new-onset diabetes mellitus in Asian population : three-year clinical outcomes", Int J Cardiol, 184, pp.502~506 (2015)
- Rha SW, "Impact of statin use on development of new-onset diabetes mellitus in Asian population", Am J Cardiol, 117, pp.382~387 (2016)
- Almdal T, "The independent effect of type 2 diabetes mellitus on ischemic heart disease, stroke, and death : a population-based study of 13, 000 men and women with 20 years of follow-up", Arch Intern Med, 164, pp.1422~1426 (2004)
- Wilson PW, "Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus", Circulation, 112, pp.3066~3072 (2005)
- Abbasi A, "Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study", BMJ, 345, pp.e5900~ (2012)
- Schmidt MI, "Identifying individuals at high risk for diabetes : The Atherosclerosis Risk in Communities study", Diabetes Care, 28, pp.2013~2018 (2005)
- Waljee AK, "Machine learning in medicine : a primer for physicians", Am J Gastroenterol, 105, pp.1224~1226 (2010)
- Obermeyer Z, "Predicting the future-big data, machine learning, and clinical medicine", N Engl J Med, 375, pp.1216~1219 (2016)
- Deo RC, "Machine learning in medicine", Circulation, 132, pp.1920~1930 (2015)
- Darcy AM, "Machine learning and the profession of medicine", JAMA, 315, pp.551~552 (2016)
- "Executive summary of the Third Report of the National Cholesterol Education Program(NCEP)Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults(Adult Treatment Panel III)", JAMA, 285, pp.2486~2497 (2001)
- Lai SW, "Epidemiology of fatty liver in a hospitalbased study in Taiwan", South Med J, 95, pp.1288~1292 (2002)
- Poggio ED, "Performance of the modification of diet in renal disease and Cockcroft-Gault equations in the estimation of GFR in health and in chronic kidney disease", J Am Soc Nephrol, 16, pp.459~466 (2005)
- Hall MA, "Benchmarking attribute selection techniques for discrete class data mining", IEEE Trans Knowl Data Eng, 15, pp.1437~1447 (2003)
- Motwani M, "Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease : a 5-year multicentre prospective registry analysis", Eur Heart J, 38, pp.500~507 (2017)
- Kotsiantis SB, "Supervised machine learning : a review of classification techniques", Informatica, 31, pp.249~268 (2007)
- Aha DW, "Instance-based learning algorithms", Machine Learning, 6, pp.37~66 (1991)
- Molinaro AM, "Prediction error estimation : a comparison of resampling methods", Bioinformatics, 21, pp.3301~3307 (2005)
- Witten IH, "Data mining: practical machine learning tools and techniques" (2016), Morgan Kaufmann
- Bradley AP, "The use of the area under the ROC curve in the evaluation of machine learning algorithms", Pattern Recogn, 30, pp.1145~1159 (1997)
- Gulshan V, "Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs", JAMA, 316, pp.2402~2410 (2016)