Table 2 Performance of the ConvNetHUP and ConvNetUHCMC/CWRU classifiers on the CINJ data cohort in terms of means and standard deviation of Dice coefficient, PPV and NPV.

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

Group

N

Dice

PPV

NPV

ConvNetHUP

 All cases

40

0.6771 ± 0.2445

0.6464 ± 0.2870

0.9709 ± 0.0350

 Only invasive

19

0.7578 ± 0.2166

0.7462 ± 0.2480

0.9654 ± 0.0355

 Mixture

21

0.6041 ± 0.2501

0.5560 ± 0.2953

0.5560 ± 0.2953

ConvNetUHCMC/CWRU

 All cases

40

0.6596 ± 0.2527

0.6370 ± 0.2941

0.9663 ± 0.0421

 Only invasive

19

0.7596 ± 0.2074

0.7499 ± 0.2423

0.9614 ± 0.0440

 Mixture

21

0.5691 ± 0.2602

0.5348 ± 0.3045

0.9708 ± 0.0409

  1. The results in Table 2 are organized in terms of all cases in the CINJ cohort (N = 40), a subset of the CINJ cohort with invasive breast cancer alone (N = 19), and a mixture of invasive and other in situ subtypes of breast cancer (N = 21).
Back to article page