<p>The combined model which includes sex, age, systolic blood pressure, waist, total cholesterol, HDL-C, triglycerides and <i>HbA1c</i> and 287 individual lipid species as features in the model attained a maximum AUC of 0.826 using 23 features the most frequent of which are listed in <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0076577#pone-0076577-t004" target="_blank">Table 4</a>. Solid lines denote the mean AUC and broken lines represent the corresponding 95% confidence intervals plotted against the number of features incorporated into the model. The plot is truncated at 23 features.</p
<p>The graphs only show glycaemic biomarkers (FPG and HbA1c), and the non-glycaemic biomarker, alani...
<p>AUC is the area under the receiver-operator characteristic curve. Odds ratios are displayed as od...
<p>Figures show ROC curves for a model including BMI and a model including BMI and Rib/height ratio....
<p>AUC, area under the receiver operating characteristic curve; CI, confidence interval; HDL, high-d...
<p>The area under the receiver operating characteristic curve (AUC) for clinical characteristics (ag...
<p>ALT, alanine aminotransferase; AST, asparate aminotransferase; AUC, area under the receiver opera...
<p>Bars represent AUC values. All models exclude subjects with type 2 diabetes. Statistical differen...
a<p>Risk factor model.</p>b<p>Lipid species model.</p>c<p>Combined risk factor and lipid species mod...
<p>AUC, Area under the curve; All biomarker levels were sex-specific except for hsCRP;</p><p>CDP: Se...
<p>Each ROC curve displays the sensitivity versus one minus specificity for each logistic regression...
<p>Receiver operating characteristic (ROC) curve plots for TG-AUC predicting diabetes risk.</p
<div><p>The AUC in an established conventional model was 0.753 (95% CI, 0.679 to 0.827), and that in...
<p>The 95% confidence interval (CI) is given in parentheses. AUC - area under the curve.</p
1<p>Adjusted for diabetes, hypertension, dyslipidemia, smoking, alcohol consumption, physical activi...
<p>The figure shows an ROC curve for HbA<sub>1c</sub> (continuous) to discriminate subjects with typ...
<p>The graphs only show glycaemic biomarkers (FPG and HbA1c), and the non-glycaemic biomarker, alani...
<p>AUC is the area under the receiver-operator characteristic curve. Odds ratios are displayed as od...
<p>Figures show ROC curves for a model including BMI and a model including BMI and Rib/height ratio....
<p>AUC, area under the receiver operating characteristic curve; CI, confidence interval; HDL, high-d...
<p>The area under the receiver operating characteristic curve (AUC) for clinical characteristics (ag...
<p>ALT, alanine aminotransferase; AST, asparate aminotransferase; AUC, area under the receiver opera...
<p>Bars represent AUC values. All models exclude subjects with type 2 diabetes. Statistical differen...
a<p>Risk factor model.</p>b<p>Lipid species model.</p>c<p>Combined risk factor and lipid species mod...
<p>AUC, Area under the curve; All biomarker levels were sex-specific except for hsCRP;</p><p>CDP: Se...
<p>Each ROC curve displays the sensitivity versus one minus specificity for each logistic regression...
<p>Receiver operating characteristic (ROC) curve plots for TG-AUC predicting diabetes risk.</p
<div><p>The AUC in an established conventional model was 0.753 (95% CI, 0.679 to 0.827), and that in...
<p>The 95% confidence interval (CI) is given in parentheses. AUC - area under the curve.</p
1<p>Adjusted for diabetes, hypertension, dyslipidemia, smoking, alcohol consumption, physical activi...
<p>The figure shows an ROC curve for HbA<sub>1c</sub> (continuous) to discriminate subjects with typ...
<p>The graphs only show glycaemic biomarkers (FPG and HbA1c), and the non-glycaemic biomarker, alani...
<p>AUC is the area under the receiver-operator characteristic curve. Odds ratios are displayed as od...
<p>Figures show ROC curves for a model including BMI and a model including BMI and Rib/height ratio....