A novel radiomics-based risk assessment model significantly improves prediction of early postoperative recurrence in locally advanced gastric cancer, crucial for surgical decision-making.

• AUC values range from 0.850 to 0.873 across multiple validation cohorts.
• The model outperforms traditional staging in predicting five-year overall survival (HR range: 1.830-2.166, p < 0.001).

Enhanced immune infiltration in low-risk patients offers insights into tumor biology, aiding personalized care strategies.

• Decision curve analyses highlight the model’s added clinical benefits over existing approaches.