Table 7 Stratified 10-fold cross-validation results. Global accuracy (ACC) with per-class precision (PRE), sensitivity (SEN), specificity (SPE), and area under the curve (AUC) are shown for models A and B; and contrasted with those reported by [9]. Each entry of our results corresponds to the mean value obtained across the 10 folds. Only best performing algorithms from bootstrap validation were included (i.e. those appearing in bold from Table 6)
From: Machine learning based study for the classification of Type 2 diabetes mellitus subtypes
Models A | Models B | [9] | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Scheme | Class | ACC | PRE | SEN | SPE | AUC | ACC | PRE | SEN | SPE | AUC | ACC | PRE | SEN | SPE | AUC |
S1 | MARD | 0.990 | 0.991 | 0.993 | 0.994 | 1.000 | 0.982 | 0.989 | 0.989 | 0.992 | 0.999 | 0.981 | 1.000 | 0.987 | 1.000 | 1.000 |
MORD | 0.983 | 0.985 | 0.996 | 1.000 | 0.969 | 0.986 | 0.991 | 0.998 | 0.992 | 1.000 | 0.973 | 1.000 | ||||
SIDD | 0.991 | 0.987 | 0.998 | 1.000 | 0.982 | 0.969 | 0.996 | 0.999 | 0.998 | 0.994 | 0.991 | 0.990 | ||||
SIRD | 0.994 | 0.992 | 0.999 | 1.000 | 0.985 | 0.975 | 0.997 | 1.000 | 0.998 | 1.000 | 0.991 | 1.000 | ||||
S2 | MARD | 0.832 | 0.824 | 0.894 | 0.863 | 0.945 | 0.988 | 0.990 | 0.996 | 0.992 | 1.000 | 0.903 | 0.945 | 0.876 | 0.914 | 0.880 |
MORD | 0.863 | 0.875 | 0.967 | 0.985 | 0.987 | 0.978 | 0.996 | 0.999 | 0.988 | 0.966 | 0.954 | 0.930 | ||||
SIDD | 0.901 | 0.904 | 0.975 | 0.994 | 0.986 | 0.990 | 0.997 | 0.999 | 0.984 | 0.988 | 0.931 | 0.970 | ||||
SIRD | 0.733 | 0.582 | 0.949 | 0.907 | 0.989 | 0.982 | 0.998 | 1.000 | 0.901 | 0.959 | 0.656 | 0.860 | ||||
S3 | MARD | 0.760 | 0.734 | 0.902 | 0.764 | 0.898 | 0.899 | 0.937 | 0.964 | 0.951 | 0.990 | 0.859 | 0.912 | 0.856 | 0.891 | 0.840 |
MORD | 0.804 | 0.809 | 0.953 | 0.954 | 0.898 | 0.903 | 0.970 | 0.989 | 0.979 | 0.950 | 0.921 | 0.920 | ||||
SIDD | 0.814 | 0.792 | 0.955 | 0.958 | 0.891 | 0.940 | 0.975 | 0.994 | 0.935 | 0.954 | 0.774 | 0.850 | ||||
SIRD | 0.716 | 0.372 | 0.965 | 0.891 | 0.789 | 0.677 | 0.965 | 0.942 | 0.903 | 0.949 | 0.627 | 0.840 | ||||
S4 | MARD | 0.979 | 0.985 | 0.991 | 0.989 | 0.999 | 0.985 | 0.990 | 0.993 | 0.992 | 0.999 | – | – | – | – | – |
MORD | 0.970 | 0.964 | 0.993 | 0.999 | 0.971 | 0.982 | 0.992 | 0.998 | – | – | – | – | ||||
SIDD | 0.981 | 0.978 | 0.995 | 0.999 | 0.994 | 0.974 | 0.999 | 0.999 | – | – | – | – | ||||
SIRD | 0.974 | 0.968 | 0.994 | 0.998 | 0.982 | 0.982 | 0.997 | 1.000 | – | – | – | – | ||||
S5 | MARD | 0.833 | 0.824 | 0.877 | 0.865 | 0.943 | 0.987 | 0.992 | 0.994 | 0.994 | 0.999 | – | – | – | – | – |
MORD | 0.874 | 0.904 | 0.969 | 0.985 | 0.973 | 0.991 | 0.992 | 0.999 | – | – | – | – | ||||
SIDD | 0.910 | 0.922 | 0.977 | 0.994 | 0.994 | 0.978 | 0.999 | 0.999 | – | – | – | – | ||||
SIRD | 0.714 | 0.578 | 0.945 | 0.906 | 0.991 | 0.976 | 0.998 | 1.000 | – | – | – | – | ||||
S6 | MARD | 0.825 | 0.829 | 0.861 | 0.873 | 0.948 | 0.983 | 0.987 | 0.992 | 0.990 | 0.999 | – | – | – | – | – |
MORD | 0.857 | 0.878 | 0.965 | 0.984 | 0.977 | 0.974 | 0.993 | 0.998 | – | – | – | – | ||||
SIDD | 0.914 | 0.912 | 0.979 | 0.994 | 0.984 | 0.981 | 0.996 | 0.999 | – | – | – | – | ||||
SIRD | 0.678 | 0.605 | 0.930 | 0.911 | 0.979 | 0.972 | 0.996 | 1.000 | – | – | – | – | ||||
S7 | MARD | 0.738 | 0.715 | 0.904 | 0.738 | 0.881 | 0.982 | 0.991 | 0.990 | 0.993 | 0.999 | 0.820 | 0.923 | 0.878 | 0.884 | 0.880 |
MORD | 0.797 | 0.819 | 0.949 | 0.957 | 0.975 | 0.970 | 0.992 | 0.998 | 0.981 | 0.971 | 0.934 | 0.940 | ||||
SIDD | 0.880 | 0.881 | 0.971 | 0.980 | 0.971 | 0.976 | 0.994 | 0.999 | 0.984 | 0.979 | 0.928 | 0.950 | ||||
SIRD | 0.404 | 0.153 | 0.943 | 0.753 | 0.980 | 0.981 | 0.996 | 0.999 | 0.897 | 0.941 | 0.601 | 0.820 | ||||