Multiparametric Magnetic Resonance Imaging Information Fusion Using Graph Convolutional Network for Glioma Grading

Table 3

The performance of three methods of fusing mpMRI context and multiparameter information in two different datasets is compared.

Dataset	Method	Base-ResNet
		AUC	ACC
BraTS2020	N(1)	0.949	0.920
	N(2)	0.970	0.920
	MMIF-GCN	0.986	0.947
GliomaHPPH2018	N(1)	0.965	0.894
	N(2)	0.969	0.936
	MMIF-GCN	0.996	0.979

The results of our study are shown in bold.