Skip to main content
. 2022 Oct 14;12:974467. doi: 10.3389/fonc.2022.974467

Table 4.

The training and testing AUC between different feature robustness pre-screening thresholds.

Outcomes ICC > 0 ICC > 0.75 ICC > 0.95
Training AUC Testing AUC Training AUC Testing AUC Training AUC Testing AUC
DM HN1 0.96 0.52 0.92 0.53 0.82 0.60
HN-PETCT 0.84 0.69 0.82 0.70 0.74 0.70
HNSCC 0.76 0.53 0.68 0.50 0.63 0.53
OPC 0.72 0.60 0.68 0.62 0.64 0.62
LR HN1 0.86 0.57 0.82 0.60 0.70 0.60
HN-PETCT 0.83 0.62 0.79 0.63 0.70 0.54
HNSCC 0.74 0.62 0.72 0.64 0.68 0.65
OPC 0.72 0.52 0.69 0.54 0.61 0.54