Table 1 GasHisSDB dataset distribution description.

From: An interpretable framework for gastric cancer classification using multi-channel attention mechanisms and transfer learning approach on histopathology images

Sub-dataset

Cropping size

Normal

Abnormal

Sub-dataset A

160 x 160

20,160

13,124

Sub-dataset B

120 x 120

40,460

24,801

Sub-dataset C

80 x 80

87,500

59,151

Total

 

148,120

97,076

Back to article page