Table 4 Classification results for a ResNet-50 fine-tuned on mouth, esophagus, stomach, small intestine, and colon.

From: Galar - a large multi-label video capsule endoscopy dataset

Label

F1

AUROC

Accuracy

mouth

0.42

1.00

0.75

esophagus

0.65

1.00

0.73

stomach

0.78

0.93

0.93

small intestine

0.93

0.95

0.92

colon

0.75

0.96

0.72

macro average

0.71

0.96

0.81

micro average

0.89

—

0.89

  1. The metrics are computed individually for each label, and both macro- and micro-averaged scores are calculated across all labels. The outcomes are averaged across the 5 cross-validation folds.
Back to article page