Table 1. The composition of training (6000 images), validation (20 images), and test (13 images) datasets.

	Subset 1	Subset 2	Subset 3	Subset 4	Subset 5
Fold 1	Validation	Train	Train	Train	Test
Fold 2	Train	Validation	Train	Train	Test
Fold 3	Train	Train	Validation	Train	Test
Fold 4	Train	Train	Train	Validation	Test