Skip to main content

Table 2 Normalized confusion matrix of automated image segmentation on the synthesized PR-stained cross section images. The overall prediction accuracy measured by the sum of diagonal elements was 0.9831.

From: Automated recognition of cell phenotypes in histology images based on membrane- and nuclei-targeting biomarkers

True class

Predicted class

 

P1

P2

P3

P4

P5

total

P1

0.3460

0.0038

0.0000

0.0000

0.0000

0.3498

P2

0.0013

0.2886

0.0016

0.0003

0.0000

0.2919

P3

0.0003

0.0066

0.3068

0.0000

0.0005

0.3142

P4

0.0000

0.0012

0.0000

0.0200

0.0000

0.0212

P5

0.0000

0.0000

0.0012

0.0000

0.0217

0.0230

Total

0.3476

0.3002

0.3096

0.0203

0.0222

1.0000

Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us