AUG 14, 2026 · JOURNAL ARTICLE
Building an interpretable machine learning prognosis prediction model-based on baseline examinations of patients with esophageal cancer undergoing surgery
Frontiers in Oncology
Single-centre retrospective registry study developing a machine learning prediction model with moderate discrimination (AUC 0.69) on a private dataset; lacks external validation, comparison to standard prognostic tools, and prospective clinical outcome assessment.
Reported
Accuracy (1-year)0.74
F1 score (1-year)0.72
Area Under the Curve0.69