Skip to main content
Fig. 5 | Cardiovascular Diabetology

Fig. 5

From: Machine-learning to stratify diabetic patients using novel cardiac biomarkers and integrative genomics

Fig. 5

Feature importance of best factors combined from patients. The most important predictive parameters using a binary and b multiple classification with HbA1c, the absolute value of a feature being high (red) or low (blue) depicting diabetic (right-side) or non-diabetic (left-side) status. The most important predictive parameters using c binary and d multiple classification without HbA1c, the absolute value of a feature being high (red) or low (blue) depicting diabetic (right-side) or non-diabetic (left-side) status. SHAP e binary and f multiple classification depiction without HbA1c of the interaction between total nuclear methylation and CpG24 methylation. SHAP values > 0.0 are diabetic (T2DM), SHAP values < 0.0 are non-diabetic (ND), SHAP values = 0 are either ND or T2DM without influence on the model. ND: non-diabetic; T2DM: type 2 diabetic; HbA1c: glycated hemoglobin; CpG: cytosine nucleotide followed by a guanine nucleotide; Nuc: nuclear; 5mC: 5-methylcytosine; binary: no diabetes and diabetes; multiple: no diabetes, prediabetes, and type 2 diabetes

Back to article page