Prediction of metabolic syndrome using artificial neural network system based on clinical data including insulin resistance index and serum adiponectin

Hiroshi Hirose, Tetsuro Takayama, Shigenari Hozawa, Toshifumi Hibi, Ikuo Saito

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

Objective: This study aimed to predict the 6-year incidence of metabolic syndrome (MetS) using an artificial neural network (ANN) system and multiple logistic regression (MLR) analysis based on clinical factors, including the insulin resistance index calculated by homeostasis model assessment (HOMA-IR). Design: Subjects were recruited from participants in annual health check-ups in both 2000 and 2006. A total of 410 Japanese male teachers and other workers at Keio University, 30-59 years of age at baseline, participated in this retrospective cohort study. Measurements: Clinical parameters were randomly divided into a training dataset and a validation dataset, and the ANN system and MLR analysis were applied to predict individual incidences. The leave some out cross validation method was used for validation. Results: The sensitivity of the prediction was 0.27 for the MLR model and 0.93 for the ANN system, while specificities were 0.95 and 0.91, respectively. Sensitivity analysis employing the ANN system identified BMI, age, diastolic blood pressure, HDL-cholesterol, LDL-cholesterol and HOMA-IR as important predictors, suggesting these factors to be non-linearly related to the outcome. Conclusion: We successfully predicted the 6-year incidence of MetS using an ANN system based on clinical data, including HOMA-IR and serum adiponectin, in Japanese male subjects.

Original languageEnglish
Pages (from-to)1051-1056
Number of pages6
JournalComputers in Biology and Medicine
Volume41
Issue number11
DOIs
Publication statusPublished - 2011 Nov

Fingerprint

Insulin
Adiponectin
Insulin Resistance
Logistic Models
Neural networks
Logistics
Serum
Cholesterol
Incidence
Regression analysis
Regression Analysis
Blood Pressure
LDL Cholesterol
HDL Cholesterol
Blood pressure
Homeostasis
Cohort Studies
Retrospective Studies
Sensitivity analysis
Health

Keywords

  • Artificial neural network system
  • Insulin resistance index
  • Metabolic syndrome

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

Prediction of metabolic syndrome using artificial neural network system based on clinical data including insulin resistance index and serum adiponectin. / Hirose, Hiroshi; Takayama, Tetsuro; Hozawa, Shigenari; Hibi, Toshifumi; Saito, Ikuo.

In: Computers in Biology and Medicine, Vol. 41, No. 11, 11.2011, p. 1051-1056.

Research output: Contribution to journalArticle

@article{66fe55b1b55c47619e0d3d27666554b8,
title = "Prediction of metabolic syndrome using artificial neural network system based on clinical data including insulin resistance index and serum adiponectin",
abstract = "Objective: This study aimed to predict the 6-year incidence of metabolic syndrome (MetS) using an artificial neural network (ANN) system and multiple logistic regression (MLR) analysis based on clinical factors, including the insulin resistance index calculated by homeostasis model assessment (HOMA-IR). Design: Subjects were recruited from participants in annual health check-ups in both 2000 and 2006. A total of 410 Japanese male teachers and other workers at Keio University, 30-59 years of age at baseline, participated in this retrospective cohort study. Measurements: Clinical parameters were randomly divided into a training dataset and a validation dataset, and the ANN system and MLR analysis were applied to predict individual incidences. The leave some out cross validation method was used for validation. Results: The sensitivity of the prediction was 0.27 for the MLR model and 0.93 for the ANN system, while specificities were 0.95 and 0.91, respectively. Sensitivity analysis employing the ANN system identified BMI, age, diastolic blood pressure, HDL-cholesterol, LDL-cholesterol and HOMA-IR as important predictors, suggesting these factors to be non-linearly related to the outcome. Conclusion: We successfully predicted the 6-year incidence of MetS using an ANN system based on clinical data, including HOMA-IR and serum adiponectin, in Japanese male subjects.",
keywords = "Artificial neural network system, Insulin resistance index, Metabolic syndrome",
author = "Hiroshi Hirose and Tetsuro Takayama and Shigenari Hozawa and Toshifumi Hibi and Ikuo Saito",
year = "2011",
month = "11",
doi = "10.1016/j.compbiomed.2011.09.005",
language = "English",
volume = "41",
pages = "1051--1056",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier Limited",
number = "11",

}

TY - JOUR

T1 - Prediction of metabolic syndrome using artificial neural network system based on clinical data including insulin resistance index and serum adiponectin

AU - Hirose, Hiroshi

AU - Takayama, Tetsuro

AU - Hozawa, Shigenari

AU - Hibi, Toshifumi

AU - Saito, Ikuo

PY - 2011/11

Y1 - 2011/11

N2 - Objective: This study aimed to predict the 6-year incidence of metabolic syndrome (MetS) using an artificial neural network (ANN) system and multiple logistic regression (MLR) analysis based on clinical factors, including the insulin resistance index calculated by homeostasis model assessment (HOMA-IR). Design: Subjects were recruited from participants in annual health check-ups in both 2000 and 2006. A total of 410 Japanese male teachers and other workers at Keio University, 30-59 years of age at baseline, participated in this retrospective cohort study. Measurements: Clinical parameters were randomly divided into a training dataset and a validation dataset, and the ANN system and MLR analysis were applied to predict individual incidences. The leave some out cross validation method was used for validation. Results: The sensitivity of the prediction was 0.27 for the MLR model and 0.93 for the ANN system, while specificities were 0.95 and 0.91, respectively. Sensitivity analysis employing the ANN system identified BMI, age, diastolic blood pressure, HDL-cholesterol, LDL-cholesterol and HOMA-IR as important predictors, suggesting these factors to be non-linearly related to the outcome. Conclusion: We successfully predicted the 6-year incidence of MetS using an ANN system based on clinical data, including HOMA-IR and serum adiponectin, in Japanese male subjects.

AB - Objective: This study aimed to predict the 6-year incidence of metabolic syndrome (MetS) using an artificial neural network (ANN) system and multiple logistic regression (MLR) analysis based on clinical factors, including the insulin resistance index calculated by homeostasis model assessment (HOMA-IR). Design: Subjects were recruited from participants in annual health check-ups in both 2000 and 2006. A total of 410 Japanese male teachers and other workers at Keio University, 30-59 years of age at baseline, participated in this retrospective cohort study. Measurements: Clinical parameters were randomly divided into a training dataset and a validation dataset, and the ANN system and MLR analysis were applied to predict individual incidences. The leave some out cross validation method was used for validation. Results: The sensitivity of the prediction was 0.27 for the MLR model and 0.93 for the ANN system, while specificities were 0.95 and 0.91, respectively. Sensitivity analysis employing the ANN system identified BMI, age, diastolic blood pressure, HDL-cholesterol, LDL-cholesterol and HOMA-IR as important predictors, suggesting these factors to be non-linearly related to the outcome. Conclusion: We successfully predicted the 6-year incidence of MetS using an ANN system based on clinical data, including HOMA-IR and serum adiponectin, in Japanese male subjects.

KW - Artificial neural network system

KW - Insulin resistance index

KW - Metabolic syndrome

UR - http://www.scopus.com/inward/record.url?scp=82655187449&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=82655187449&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2011.09.005

DO - 10.1016/j.compbiomed.2011.09.005

M3 - Article

C2 - 22000697

AN - SCOPUS:82655187449

VL - 41

SP - 1051

EP - 1056

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

IS - 11

ER -