Skip to main content
. 2021 Oct 20;12:744557. doi: 10.3389/fgene.2021.744557

TABLE 2.

Performance comparison of all models.

Method AUC [95% CI] Accuracy Sensitivity Specificity Youden’s index
PRS 0.56 [0.42; 0.81] 0.60 0.62 0.56 0.21
Deep learning 0.67 [0.47; 0.83] 0.60 0.42 0.88 0.29
LASSO w/feature selection 0.85 [0.72; 0.96] 0.81 0.81 0.80 0.61
LASSO w/o feature selection 0.51 [0.39; 0.63] 0.62 0.87 0.09 0.12