Real-time use of artificial intelligence for diagnosing early gastric cancer by magnifying image-enhanced endoscopy: a multicenter diagnostic study (with videos)

Development and validation of a feature extraction-based logical anthropomorphic diagnostic system for early gastric cancer: A case-control study

“…In this study, the performance of ENDOANGEL-LA was better than sole DCNN using the same training set and comparable with ENDOANGEL-ME using 4667 M-IEE images. 16 The feature indexes related to EGC were determined in advance by experts, which may also present in non-cancerous lesions. Based on the given diagnosis logic, the model did not require a large number of EGC images to learn and summarize the difference between EGC and non-cancer.…”

“…Therefore, we retrospectively collected 4667 images (EGC, 1950; noncancerous, 2717) from 1811 patients (EGC, 1042; noncancerous lesions, 769) between Nov 15, 2016 and May 7, 2019, which were derived from a database of our previous study. 16 To develop ENDOANGEL-LA, the images viewed at full magnification under M-IEE were included, and the images where the MS and MV were difficult to observe due to blurry, reflection, out of focus, etc. were excluded.…”

“…In addition, another sole DL model named ENDOANGEL-ME (magnifying endoscopy) was also used for the comparison, which was constructed from our previous studies using 4667 M-IEE images. 16 …”

Artificial intelligence in endoscopy: transition from academic development to integration into daily endoscopic practice

“…Given its subtle endoscopic appearance and the corresponding diagnostic challenge, early gastric cancer is a logical application of AI. Several systems have recently been described, mainly focusing on primary detection systems [1][2][3].…”

“…Given its subtle endoscopic appearance and the corresponding diagnostic challenge, early gastric cancer is a logical application of AI. Several systems have recently been described, mainly focusing on primary detection systems [1][2][3].In this issue of Endoscopy, Niikura and colleagues present results of a retrospective study in which they evaluated a previously trained computer-aided detection system for early gastric cancer [4]. For this evaluation, they applied an interesting experimental set-up using stratified matching of retrospectively collected imagery in a noninferiority comparison of AI versus expert assessment.…”