Table 1.

The AUC of models predicting lymph node metastasis.

Models of lymph node metastasis prediction	Training set (AUC; 95% CI)	Test set (AUC; 95% CI)
gene signature model	0.7649; 0.6614-0.8683	0.5891; 0.386-0.7921
clinical feature model	0.7; 0.5826-0.8174	0.6619; 0.447-0.8769
radiomics signature model	0.7147; 0.6003-0.8291	0.6721; 0.4787-0.8654
gene signature model + clinical feature model	0.8247; 0.7323-0.9172	0.6235; 0.417-0.83
gene signature model + radiomics signature model	0.828; 0.7362-0.9197	0.6356; 0.4361-0.8351
clinical feature model + radiomics signature model	0.757; 0.6502-0.8638	0.6559; 0.45-0.8617
multi-feature integration nomogram model	0.8695; 0.7909-0.9482	0.6032; 0.4005-0.8059