Table 5.

Comparison with others methods on the same dataset in 10-fold cross validation.

Method	Accuracy	Sensitivity	Specificity	AUC
MRI biomarker in Moradi et al., 2015	74.7%	88.9%	51.6%	76.6%
Global grading biomarker in Tong et al., 2017	78.9%	76.0%	82.9%	81.3%
Proposed method	79.5%	86.1%	68.8%	83.6%

The performances of MRI biomarker and global grading biomarker are described in Moradi et al. (2015) and (Tong et al., 2017). The results are averages over 100 runs, and the standard deviation/confidence intervals of accuracy and AUC of the proposed method are 1.19%/[0.7922, 0.7968] and 0.83%/[0.8358, 0.8391]. Bold values indicate the best performance in each column.