Prediction of Nephropathy in Type 2 Diabetes

An Analysis of the ACCORD Trial Applying Machine Learning Techniques

Violeta Rodriguez-Romero, Richard F. Bergstrom, Brian Decker, Gezim Lahu, Majid Vakilynejad, Robert Bies

Research output: Contribution to journalArticle

Abstract

Applying data mining and machine learning (ML) techniques to clinical data might identify predictive biomarkers for diabetic nephropathy (DN), a common complication of type 2 diabetes mellitus (T2DM). A retrospective analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial was intended to identify such factors using ML. The longitudinal data were stratified by time after patient enrollment to differentiate early and late predictors. Our results showed that Random Forest and Simple Logistic Regression methods exhibited the best performance among the evaluated algorithms. Baseline values for glomerular filtration rate (GFR), urinary creatinine, urinary albumin, potassium, cholesterol, low-density lipoprotein, and urinary albumin to creatinine ratio were identified as DN predictors. Early predictors were the baseline values of GFR, systolic blood pressure, as well as fasting plasma glucose (FPG) and potassium at month 4. Changes per year in GFR, FPG, and triglycerides were recognized as predictors of late development. In conclusion, ML-based methods successfully identified predictive factors for DN among patients with T2DM.

Original languageEnglish (US)
JournalClinical and translational science
DOIs
StatePublished - Jan 1 2019

Fingerprint

Diabetic Nephropathies
Medical problems
Glomerular Filtration Rate
Type 2 Diabetes Mellitus
Learning systems
Albumins
Fasting
Creatinine
Potassium
Blood Pressure
Plasmas
Glucose
Data Mining
Blood pressure
Biomarkers
Diabetes Complications
LDL Cholesterol
Data mining
Logistics
Triglycerides

ASJC Scopus subject areas

  • Neuroscience(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

Prediction of Nephropathy in Type 2 Diabetes : An Analysis of the ACCORD Trial Applying Machine Learning Techniques. / Rodriguez-Romero, Violeta; Bergstrom, Richard F.; Decker, Brian; Lahu, Gezim; Vakilynejad, Majid; Bies, Robert.

In: Clinical and translational science, 01.01.2019.

Research output: Contribution to journalArticle

@article{e95969cdeb464a8fb9b7f2feabadfb1c,
title = "Prediction of Nephropathy in Type 2 Diabetes: An Analysis of the ACCORD Trial Applying Machine Learning Techniques",
abstract = "Applying data mining and machine learning (ML) techniques to clinical data might identify predictive biomarkers for diabetic nephropathy (DN), a common complication of type 2 diabetes mellitus (T2DM). A retrospective analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial was intended to identify such factors using ML. The longitudinal data were stratified by time after patient enrollment to differentiate early and late predictors. Our results showed that Random Forest and Simple Logistic Regression methods exhibited the best performance among the evaluated algorithms. Baseline values for glomerular filtration rate (GFR), urinary creatinine, urinary albumin, potassium, cholesterol, low-density lipoprotein, and urinary albumin to creatinine ratio were identified as DN predictors. Early predictors were the baseline values of GFR, systolic blood pressure, as well as fasting plasma glucose (FPG) and potassium at month 4. Changes per year in GFR, FPG, and triglycerides were recognized as predictors of late development. In conclusion, ML-based methods successfully identified predictive factors for DN among patients with T2DM.",
author = "Violeta Rodriguez-Romero and Bergstrom, {Richard F.} and Brian Decker and Gezim Lahu and Majid Vakilynejad and Robert Bies",
year = "2019",
month = "1",
day = "1",
doi = "10.1111/cts.12647",
language = "English (US)",
journal = "Clinical and Translational Science",
issn = "1752-8054",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - Prediction of Nephropathy in Type 2 Diabetes

T2 - An Analysis of the ACCORD Trial Applying Machine Learning Techniques

AU - Rodriguez-Romero, Violeta

AU - Bergstrom, Richard F.

AU - Decker, Brian

AU - Lahu, Gezim

AU - Vakilynejad, Majid

AU - Bies, Robert

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Applying data mining and machine learning (ML) techniques to clinical data might identify predictive biomarkers for diabetic nephropathy (DN), a common complication of type 2 diabetes mellitus (T2DM). A retrospective analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial was intended to identify such factors using ML. The longitudinal data were stratified by time after patient enrollment to differentiate early and late predictors. Our results showed that Random Forest and Simple Logistic Regression methods exhibited the best performance among the evaluated algorithms. Baseline values for glomerular filtration rate (GFR), urinary creatinine, urinary albumin, potassium, cholesterol, low-density lipoprotein, and urinary albumin to creatinine ratio were identified as DN predictors. Early predictors were the baseline values of GFR, systolic blood pressure, as well as fasting plasma glucose (FPG) and potassium at month 4. Changes per year in GFR, FPG, and triglycerides were recognized as predictors of late development. In conclusion, ML-based methods successfully identified predictive factors for DN among patients with T2DM.

AB - Applying data mining and machine learning (ML) techniques to clinical data might identify predictive biomarkers for diabetic nephropathy (DN), a common complication of type 2 diabetes mellitus (T2DM). A retrospective analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial was intended to identify such factors using ML. The longitudinal data were stratified by time after patient enrollment to differentiate early and late predictors. Our results showed that Random Forest and Simple Logistic Regression methods exhibited the best performance among the evaluated algorithms. Baseline values for glomerular filtration rate (GFR), urinary creatinine, urinary albumin, potassium, cholesterol, low-density lipoprotein, and urinary albumin to creatinine ratio were identified as DN predictors. Early predictors were the baseline values of GFR, systolic blood pressure, as well as fasting plasma glucose (FPG) and potassium at month 4. Changes per year in GFR, FPG, and triglycerides were recognized as predictors of late development. In conclusion, ML-based methods successfully identified predictive factors for DN among patients with T2DM.

UR - http://www.scopus.com/inward/record.url?scp=85066482965&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066482965&partnerID=8YFLogxK

U2 - 10.1111/cts.12647

DO - 10.1111/cts.12647

M3 - Article

JO - Clinical and Translational Science

JF - Clinical and Translational Science

SN - 1752-8054

ER -