He X, Wu L, Dong Z, et al. - In this study, experts introduced a deep learning-based system named ENDOANGEL-ME to diagnose early gastric cancer (EGC) in magnifying image-enhanced endoscopy (M-IEE), and findings demonstrate that ENDOANGEL-ME could be well employed in clinical setting.

• This study included 3,099 patients undergoing M-IEE.

• ENDOANGEL-ME showed a diagnostic accuracy of 88.44% and 90.49% in internal and external images, respectively, to diagnose EGC.

• In 93 internal videos, the accuracy of ENDOANGEL-ME was 90.32% for diagnosing EGC, which was significantly superior to that of senior endoscopists (70.16%±8.78).

• In 94 external videos, with the help of ENDOANGEL-ME, endoscopists demonstrated improved accuracy as well as sensitivity (85.64% vs 80.32%, 82.03% vs 67.19%).

• In real clinical practice, ENDOANGEL-ME provided a sensitivity of 92.59% (25/27) and an accuracy of 83.67% (210/251) in 194 prospective consecutive patients with 251 lesions.