Fig. 5

Personalized survival prediction by using (a) LMTLR model, (b) NMTLR model, (c) RSF model and (d) CoxPH model when randomly choosing two patients who have CVD outcome at 1.1 years and 2.3 years as an example. X axis is survival time in years, Y axis is survival probability. Hypertension patient one (red line) was diagnosed as CVD outcome at 2.3 survival years while hypertension patient two (blue line) was diagnosed as CVD outcome at 1.1 survival years. The bottom position of survival time text (1.1 and 2.3) on the pictures correspond to 50% survival probability horizontal location line. Note: CoxPH = Cox’s proportional hazard, LMTL = linear multi-task logistic regression, NMTLR = neural multi-task logistic regression, RSF = random survival forest