AI-supported decision-making after endoscopic resection of early gastric cancer: what the evidence shows

A review in Clinical Endoscopy examines how machine learning could support the decision made after endoscopic resection of early gastric cancer: proceed to additional gastrectomy with lymph node dissection, or observe. The clinical problem is real — lymph node metastasis occurs in roughly 5-10% of early gastric cancers, yet about 90% of patients sent for additional surgery turn out to have none, meaning most of those operations are avoidable. Published models built on clinicopathological variables reach AUC 0.69-0.94 and in several comparisons outperform the eCura scoring system in use today, while whole-slide-image analysis of routine H&E offers a more reproducible read, although evidence specific to gastric cancer remains limited. This is a narrative review rather than a clinical trial, and the authors frame these tools as support for the clinician's judgement, not a replacement for it.