Table 4 Performance comparison of ViT-based models with different classifiers and familiar testing data.

Model	Accuracy	Precision		Recall		F1-Score
Model	Accuracy	Tumor	No tumor	Tumor	No tumor	Tumor	No tumor
ViT + RF (Trained	0.9817	0.9707	0.9932	0.9933	0.9700	0.9819	0.9815
on CLAHE Images)	0.9817	0.9707	0.9932	0.9933	0.9700	0.9819	0.9815
ViT + SVM (Trained	0.9817	0.9833	0.9801	0.9800	0.9833	0.9816	0.9817
on CLAHE Images)	0.9817	0.9833	0.9801	0.9800	0.9833	0.9816	0.9817
ViT + RF (Trained	0.9350	0.9279	0.9424	0.9433	0.9267	0.9355	0.9345
on Contour-
Region Masked
Image)
ViT + SVM	0.9417	0.9373	0.9461	0.9467	0.9367	0.9414	0.9420
(Trained on
Contour-Region
Masked Image)
ViT + RF (Trained	0.9033	0.8929	0.9144	0.9167	0.8900	0.9046	0.9020
on HOG Images)	0.9033	0.8929	0.9144	0.9167	0.8900	0.9046	0.9020
ViT + SVM (Trained	0.9217	0.9203	0.9231	0.9233	0.9200	0.9218	0.9215
on HOG Images)	0.9217	0.9203	0.9231	0.9233	0.9200	0.9218	0.9215

Quick links

Search