Table 6.

Results of ensemble deep learning models with multiclass imbalanced dataset.

Model	Accuracy	Precision	Recall	F1 Score	AUC
EfficientNet-B2+DenseNet-121	92.82	94.29	93.76	91.52	99.38
VGG-16+DenseNet-121	91.52	95.21	92.43	92.11	98.96
EfficientNet-B2+Xception	90.45	92.61	87.80	90.40	98.80
Xception+DenseNet-121	89.29	85.83	88.90	90.87	98.28
VGG-16+EfficientNet-B2	95.00	95.23	93.34	96.13	99.41