Table 4.

Test result data on pulmonary chest X-ray datasets. Each neural network has two rows of conclusion data. The first row of data corresponds to the original neural network framework proposed by the authors. In contrast, the second row of data represents the test results after incorporating our method SAA-SDM

Name	mIoU	Accuracy	Acceleration
Unet	0.9363	0.9699	-
	0.9334	0.9685	57.14%
SegNet	0.9175	0.9606	-
	0.9324	0.9679	86.67%
UNet +  +	0.8829	0.9693	-
	0.8934	0.9683	33.33%
UNet3 +	0.8989	0.9577	-
	0.9092	0.9629	0%
U2Net	0.8654	0.9339	-
	0.8793	0.9412	11.11%
SegFormer	0.8415	0.9387	-
	0.8883	0.9583	∞
SegMenter	0.8251	0.9130	-
	0.8372	0.9192	29.41%