Table 3.

Detail of datasets used by the proposed CXAU-Net model.

Segmentation type	Multi-class Segmentation				Binary-class Segmentation
Approach used	Image-based		Patch-based		Image-based
Dataset type	Training dataset	Testing dataset	Training dataset	Testing dataset	Training dataset	Testing Dataset
Dataset	COVID-19 dataset	Chest X-ray 14 dataset	COVID-19 dataset	Chest X-ray 14 dataset	NLM-China and MC CXR datasets	JSRT dataset
Number of Images	105 images	640 images with 160 images of each case	210 patches	640 patches with 160 patches of each case	704 images	247 images
Number of Augmented Images used	840 images	–	1680 patches	–	5632 images	–
Size of images	256×256×3	256×256×3	128×128×3	128×128×3	256×256×1	256×256×1