Table 4.

Patch-wise model performance on the testing set in comparison to a previous study.

Model	Weighted average Accuracy	Weighted average Precision	Weighted average Recall	Weighted average F1 Score
ResNet50	0.68	0.69	0.68	0.68
ResNet101	0.78	0.78	0.78	0.78
ResNet101_imgnet	0.73	0.75	0.73	0.74
VGG16	0.73	0.75	0.70	0.73
Xception	0.77	0.78	0.77	0.77
Jaber et al.	0.59–0.66	N/A	N/A	N/A

Bold font denote the best model. N/A: no information available.