Skip to main content
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: Artif Intell Med. 2016 Jul 27;72:42–55. doi: 10.1016/j.artmed.2016.07.001

Figure 4.

Figure 4

The combined probability distributions for a positive biopsy, of DBN A (top) and DBN B (bottom), for all cases across the 10 random test sets, for each screen. Red indicates all confirmed cancer cases in the trial, irrespective of screening time points. Blue indicates the confirmed non-cancer cases. The three subplots depict the probability of a positive biopsy in each of the three screening points of the trial. With successive screenings we can see that the probability of a positive biopsy for non-cancer (blue) and cancer (red) cases tends to move towards the left and right side of each subplot, respectively. The solid black lines represent the thresholds chosen to discriminate cancer cases from non-cancer cases in the DBN predictions.