Predicting the magnitude of risk for non-curative endoscopic submucosal dissection in superficial esophageal cancer using explainable artificial intelligence

Key Points

• Non-curative resection risk can be predicted prior to endoscopic submucosal dissection.
• The prediction model uses logistic regression, providing effective risk identification.
• SHAP-based interpretability enhances the transparency of the risk assessment tool.
• This model highlights the need for reliable predictive analytics in esophageal cancer treatment.