Gastric cancer (GC) is a significant healthcare concern, and the identification of high-risk patients is crucial. Indeed, gastric precancerous conditions present significant diagnostic challenges, particularly early intestinal metaplasia (IM) detection. This study developed a deep learning system to assist in IM detection using image patches from gastric corpus examined using virtual chromoendoscopy in a Western country. Utilizing a retrospective dataset of endoscopic images from Sant'Andrea University Hospital of Rome, collected between January 2020 and December 2023, the system extracted 200 × 200 pixel patches, classifying them with a voting scheme. The specificity and sensitivity on the patch test set were 76% and 72%, respectively. The optimization of a learnable voting scheme on a validation set achieved a specificity of 70% and sensitivity of 100% for entire images. Despite data limitations and the absence of pre-trained models, the system shows promising results for preliminary screening in gastric precancerous condition diagnostics, providing an explainable and robust Artificial Intelligence approach.

After being ingested and entering the human stomach, Helicobacter pylori (H. pylori) adopts several effective strategies to adhere to and colonize the gastric mucosa and move to different regions of the stomach to obtain more nutrients and escape from the harsher environments of the stomach, leading to acute infection and chronic gastritis, which is the basis of malignant gastric tumors. The endoscopic manifestations and pathological features of H. pylori infection are diverse and vary with the duration of infection. In this review, we describe the endoscopic manifestations of each stage of H. pylori gastritis and then reveal the potential mechanisms of bacterial intragastric colonization and migration from the perspective of endoscopists to provide direction for future research on the effective therapy and management of H. pylori infection.

Objective: To conduct a systematic review and meta-analysis on the prediction of severity of gastric intestinal metaplasia (GIM) in localized and entire gastric mucosa using endoscopy. Methods: The authors searched Web of Science, PubMed, Embase and Cochrane Central Register of Controlled Trials and performed systematic searches on endoscopic grading of GIM of the entire stomach using Meta-DiSc and Stata. Results: Sensitivity and specificity for the stratified prediction of overall GIM were 0.91 (95% CI: 0.85-0.95) and 0.91 (95% CI: 0.88-0.93), respectively. Sensitivity in predicting the different grades of GIM was higher in operative link on GIM assessment grades 0, III and IV but lower in grades I and II. Conclusion: Digital chromoendoscopy is well suited to predicting the severity of localized and overall GIM.

Gastric intestinal metaplasia (GIM) is a precancerous lesion of the stomach, which severely affects human life and health. Currently, a variety of endoscopic techniques are used to screen/evaluate GIM. Traditional white-light endoscopy (WLE) and acetic-acid chromoendoscopy combined with magnifying endoscopy (ME-AAC) are the interventions of choice due to their diagnostic efficacy for GIM. Optical-enhanced magnifying endoscopy (ME-OE) is a new virtual chromoendoscopy technique to identify GIM, which combines bandwidth-limited light and image enhancement processing technology to enhance the detection of mucosal and vascular details. We hypothesized that ME-OE is superior to WLE and ME-AAC in the evaluation of GIM.

To directly compare the diagnostic value of WLE, ME-AAC, and ME-OE for detection of GIM.

A total of 156 patients were subjected to consecutive upper gastrointestinal endoscopy examinations using WLE, ME-AAC, and ME-OE. Histopathological findings were utilized as the reference standard. Accuracy, sensitivity, specificity, and positive and negative predictive values of the three endoscopy methods in the diagnosis of GIM were evaluated. Moreover, the time to diagnosis with ME-AAC and ME-OE was analyzed. Two experts and two non-experts evaluated the GIM images diagnosed using ME-OE, and diagnostic accuracy and intra- and inter-observer agreement were analyzed.

GIM was detected in 68 of 156 patients (43.6%). The accuracy of ME-OE was highest (91.7%), followed by ME-AAC (86.5%), while that of WLE (51.9%) was lowest. Per-site analysis showed that the overall diagnostic accuracy of ME-OE was higher than that of ME-AAC (P = 0.011) and WLE (P < 0.001). The average diagnosis time was lower in ME-OE than in ME-AAC (64 ± 7 s vs 151 ± 30 s, P < 0.001). Finally, the inter-observer agreement was strong for both experts (k = 0.862) and non-experts (k = 0.800). The internal consistency was strong for experts (k = 0.713, k = 0.724) and moderate for non-experts (k = 0.667, k = 0.598).

For endoscopists, especially experienced endoscopists, ME-OE is an efficient, convenient, and time-saving endoscopic technique that should be used for the diagnosis of GIM.