Intratumoral and peritumoral radiomics-based machine learning models for the postoperative survival prediction in esophageal squamous cell carcinoma

Key Points

• To develop machine learning models that use radiomic features for predicting postoperative survival in esophageal squamous cell carcinoma (ESCC).
• Utilized intratumoral and peritumoral radiomic features
• Applied machine learning algorithms including GBDT
• Analyzed patient data post-surgery for survival outcomes
• The GBDT model showed high effectiveness in risk stratification
• Models indicated potential for improved survival prediction for ESCC patients