Figure 3
The x 1 -x 2 scatter plot and the logistic regression with interaction boundary. This figure shows the two risk factor scatter plot for the cases (diamonds) and controls (multiplication signs) for the LR model with x1 × x2 interaction. Each point represents a given sample's (x1, x2) risk vector plotted in component. The solid line is the LR model separation boundary (solid) and the curved dashed line is the Eq. (1) (ideal) separation boundary. In comparison with Figure 2, there is a slight curvature in the boundary on the right side. The sensitivity = 0.80 and false positive fraction = 0.40