Skip to main content
Fig. 1 | Cardiovascular Diabetology

Fig. 1

From: The novel inflammatory biomarker GlycA and triglyceride-rich lipoproteins are associated with the presence of subclinical myocardial dysfunction in subjects with type 1 diabetes mellitus

Fig. 1

 A: Estimated and cross-validated ROC curve of the PLS-DA classification model. B: Contribution of each variable to the multidimensional latent variables LV1 and LV2. PLS-DA relates the X matrix (input data including NMR derived parameters, clinical and anthropometric variables) and the Y matrix (presence or absence of myocardial disfunction) to find the maximum discrimination between both groups. AUROC: area under the ROC curve, PLS-DA: partial least squares discriminant analysis, LV: latent variable, NMR: nuclear magnetic resonance

Back to article page

Cardiovascular Diabetology

ISSN: 1475-2840