Figure 2 : Example results for the ConvNetHUP classifier on the CINJ validation data cohort.

From: Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent

Figure 2

The probability map predicted by the ConvNetHUP classifier (second row, (DF)) was compared against ground truth annotations by a pathologist (first row (AC)). The third row shows the evaluation results of the ConvNetHUP classifier in terms of TP (green), FN (red), FP (yellow), and TN (blue) regions.