Prediction of genetic alterations from gastric cancer histopathology images using a fully automated deep learning approach

Table 1 Accuracy, sensitivity, specificity, and F1 score of the classification results of mutation prediction models for the The Cancer Genome Atlas datasets

	Accuracy	Sensitivity	Specificity	F1 score
TCGA Frozen Tissue Slides
CDH1	0.847	0.700	0.952	0.792
ERBB2	0.716	1.000	0.512	0.746
KRAS	0.773	0.883	0.708	0.745
PIK3CA	0.834	0.771	0.884	0.806
TP53	0.667	0.743	0.602	0.673
TCGA FFPE Tissue Slides
CDH1	0.820	0.608	1.000	0.756
ERBB2	0.574	1.000	0.285	0.655
KRAS	0.894	0.727	1.000	0.842
PIK3CA	0.803	0.629	0.923	0.723
TP53	0.673	0.678	0.668	0.659

TCGA: The Cancer Genome Atlas; FFPE: Formalin-fixed paraffin-embedded.