Table 3.

Average results for the five tested classification approaches, i.e., average of the results for the two splits.

Method	Macro Average			Micro Average
	Precision	Recall	F1-score	Precision	Recall	F1-score	MCC (RK)
Pre-Trained ResNet-50	0.589	0.536	0.530	0.839	0.839	0.839	0.826
Pre-Trained ResNet-152	0.639	0.605	0.606	0.906	0.906	0.906	0.898
Pre-Trained DensNet-161	0.640	0.616	0.619	0.907	0.907	0.907	0.899
Averaged ResNet-152 + DenseNet-161	0.633	0.615	0.617	0.910	0.910	0.910	0.902
ResNet-152 + DenseNet-161 + MLP	0.612	0.606	0.605	0.909	0.909	0.909	0.902
Random Guessing	0.044	0.038	0.034	0.044	0.044	0.044	0.000
Majority Class	0.004	0.043	0.008	0.108	0.108	0.108	N/A