Artificial intelligence models demonstrated significant predictive capability for lymph node metastasis (LNM) in patients with T1 colorectal cancer, outperforming guideline-recommended metrics. Among 1,386 analyzed patients, 12.5% exhibited LNM, with AUROC values of models including regularized logistic regression classifier (0.673) and catboost classifier (0.679) surpassing the expected 0.525. These findings suggest AI integration could enhance surgical decision-making for colorectal cancer treatments across different tumor locations and treatment modalities.
Journal Article by Baek JE, Yi H (…) Byeon JS et 10 al. in Gut Liver