Interpretable AI-driven prediction of early postoperative recurrence in locally advanced rectal cancer using the Systemic Inflammatory–Nutritional Index (SINTI): A multi-center study and clinical web tool

Key Points

• The aim is to develop an interpretable AI framework for predicting early postoperative recurrence in locally advanced rectal cancer using the Systemic Inflammatory–Nutritional Index.
• Utilized an interpretable AI framework
• Conducted a multi-center study
• Integrated the Systemic Inflammatory–Nutritional Index for predictions
• Focused on post-chemoradiotherapy (NCRT) recurrence risk
• Demonstrated potential for accurate recurrence risk prediction
• Highlighted transparency and interpretability of the AI model
• Indicated need for further prospective validation of the framework