Skip to main content
Visual Computing for Industry, Biomedicine, and Art

Table 5 AUC scores of three CM sets with nine different strides for LMD + LMS where D represents displacement

From: Multi-scale characterizations of colon polyps via computed tomographic colonography

 

D = 1

D = 2

D = 3

S = 1

0.8818 ± 0.0438

0.8725 ± 0.0461

0.8706 ± 0.0489

S = 2

0.8621 ± 0.0487

0.8486 ± 0.0458

0.8751 ± 0.0477

S = 3

0.9043 ± 0.0382

0.8765 ± 0.0407

0.8786 ± 0.0463

S = 4

0.8919 ± 0.0382

0.8834 ± 0.0368

0.8907 ± 0.0417

S = 5

0.9159 ± 0.0408

0.8971 ± 0.0443

09073 ± 0.0446

S = 6

0.9057 ± 0.0394

0.8988 ± 0.0400

09042 ± 0.0447

S = 7

0.9450 ± 0.0285

0.9303 ± 0.0393

0.9223 ± 0.0375

S = 8

0.8887 ± 0.0366

0.8924 ± 0.0436

0.8838 ± 0.0397

S = 9

08825 ± 0.0452

0.9073 ± 0.0421

0.8714 ± 0.0402

Average

0.8975 ± 0.0.399

0.8896 ± 0.0421

0.8893 ± 0.0435

  1. AUC Area under the curve of receiver operating characteristic curve, CM Co-occurrence matrix, LMS Learning model by multiple strides, LMD Learning model by multiple displacements
Back to article page