A novel endoscopic artificial intelligence system to assist in the diagnosis of autoimmune gastritis: a multicenter study

• DOI: 10.1055/a-2451-3071
• Publication Date: 2024-10-24T23:10:59Z
• Authors: Shurong Chen; Louzhe Xu; Ting Li; Yi Chen; Lingling Yan; Jie Zhang; Xuefeng Zhou; Jiayi Wang; Tianlian Yan; Jinghua Wang; Xinjue He; Han Ma; Xuequn Zhang; Shenghua Zhu; Yizhen Zhang; Chengfu Xu; Jianguo Gao; Xia Ji; Dezhi Bai; Yuan Chen; Hongda Chen; Yini Ke; Lan Li; Chaohui Yu; Xinli Mao
• Keywords: Atrophic gastritis; Scope (computer science)

Background & Aims: Autoimmune gastritis (AIG), distinct from Helicobacter pylori-associated atrophic (HpAG), is underdiagnosed due to limited awareness. This multicenter study aims develop a novel endoscopic artificial intelligence (AI) system assisting in AIG diagnosis. Methods: Patients diagnosed with AIG, as well HpAG and non-atrophic (NAG), were retrospectively enrolled six centers. Endoscopic images relevant demographic medical data, collected for the development of AI-assisted system, SEER-SCOPE AI, based on multi-site feature fusion model. The diagnostic performance AI was evaluated internal external datasets. Endoscopists’ without support tested compared using Mann-Whitney U test. Heatmap analysis performed interpret AI. Results: 1 070 patients (294 386 HpAG, 390 NAG) 18 828 endoscopy collected. achieved strong identifying 96.9% sensitivity, 92.2% specificity an AUROC 0.990 internally, 90.3% 93.1% 0.973 externally. (sensitivity 91.3%) comparable experts (87.3%) significantly outperformed non-experts (70.0%). With support, overall endoscopists improved (sensitivity: [95% CI 86.0%-93.2%] vs. 78.7% 73.6%-83.2%], p=0.008). revealed consistent focus regions corresponding areas. Conclusions: demonstrated expert-level enhanced ability endoscopists. Its application holds promise potent endoscopy-assisted tool guiding biopsy sampling early detection AIG.