Artificial Intelligence–Assisted Endoscopic Detection of Early Gastric Cancer

Abstract

Early detection of gastric cancer remains challenging during routine endoscopy. This study developed and validated a deep learning–based artificial intelligence system for real-time detection of early gastric cancer. Using over 50,000 annotated endoscopic images, the model achieved high sensitivity and specificity in lesion detection. In a prospective clinical evaluation, AI-assisted endoscopy significantly improved diagnostic accuracy compared with conventional assessment. These findings demonstrate the potential of artificial intelligence to enhance early gastric cancer detection and reduce missed diagnoses in clinical practice.