Table 4.

Combinations of Dot blot parameters for selected pairwise classification problems. Shown cutoffs are designated for the logistic model e^M/(1 + e^M).

Classification: DM versus DN Macro Model: M = –12.49 + 6.03 * APOA1 + 4.51 * B2M + 5.74 * CDH1 Cutoff: 0.50, Accuracy: 76 [68, 84], Sensitivity: 72 [59, 83], Specificity: 80 [67, 89]

Classification: DM versus DN Micro Model: M = –18.91 + 9.24 * APOA1 + 5.05 * B2M + 9.48 * CDH1 Cutoff: 0.58, Accuracy: 85 [77, 91], Sensitivity: 79 [66, 89], Specificity: 90 [79, 96]

Classification: DN Micro versus DN Macro Model: M = –1.99 + 2.801 * CDH1 -2.34 * REG1A Cutoff: 0.55, Accuracy: 69 [59, 77], Sensitivity: 53 [39, 66], Specificity: 84 [73, 93]