Out of the 191 patients, 94 (49.2%) were pathologically confirmed as VETC-positive. No significant differences in diagnostic performance were observed between CT and EOB-MRI across all criteria (all P > 0.05). However, the VETC nomogram, which incorporates tumor size, necrosis, and enhancement pattern, demonstrated superior performance compared to both the SN score and Fan's model. On CT, the AUC for the VETC nomogram was 0.825, compared to 0.716 for the SN score and 0.588 for Fan’s model. Similarly, on EOB-MRI, the VETC nomogram achieved an AUC of 0.841, while the SN score and Fan’s model had AUCs of 0.721 and 0.621, respectively (both P < 0.001). Stratification based on the VETC nomogram, validated through both CT and EOB-MRI, was significantly associated with early recurrence (P < 0.05).