No abstract

Objectives: Detecting early gastric cancer is difficult, and it may even be overlooked by experienced endoscopists. Recently, artificial intelligence based on deep learning through convolutional neural networks (CNNs) has enabled significant advancements in the field of gastroenterology. However, it remains unclear whether a CNN can outperform endoscopists. In this study, we evaluated whether the performance of a CNN in detecting early gastric cancer is better than that of endoscopists. Methods: The CNN was constructed using 13,584 endoscopic images from 2639 lesions of gastric cancer. Subsequently, its diagnostic ability was compared to that of 67 endoscopists using an independent test dataset (2940 images from 140 cases). Results: The average diagnostic time for analyzing 2940 test endoscopic images by the CNN and endoscopists were 45.5 AE 1.8 s and 173.0 AE 66.0 min, respectively. The sensitivity, specificity, and positive and negative predictive values for the CNN were 58.4%, 87.3%, 26.0%, and 96.5%, respectively. These values for the 67 endoscopists were 31.9%, 97.2%, 46.2%, and 94.9%, respectively. The CNN had a significantly higher sensitivity than the endoscopists (by 26.5%; 95% confidence interval, 14.9-32.5%). Conclusion: The CNN detected more early gastric cancer cases in a shorter time than the endoscopists. The CNN needs further training to achieve higher diagnostic accuracy. However, a diagnostic support tool for gastric cancer using a CNN will be realized in the near future.

No abstract

No abstract

Background and study aim: There is no enough data for endoscopic resection (ER) of superficial duodenal epithelial tumors (SDETs) due to its rarity. There are two main kinds of ER techniques for SDETs: EMR and ESD. In addition, modified EMR techniques, underwater EMR (UEMR) and cold polypectomy (CP), are getting popular. We conducted a large-scale retrospective multicenter study to clarify detailed outcomes of duodenal ER. Patients and methods : Patients with SDETs who underwent ER at 18 institutions from January 2008 to December 2018 were included. The rates of en bloc resection and delayed adverse events (AEs) (defined as delayed bleeding or perforation) were analyzed. Local recurrence was analyzed using Kaplan-Meier method. Results: In total, 3107 patients (including 1017 receiving ESD) were included. En bloc resection rates were 79.1%, 78.6%, 86.8%, and 94.8%, and delayed AE rates were 0.5%, 2.2%, 2.8%, and 7.3% for CP, UEMR, EMR and ESD, respectively. The delayed AE rate was significantly higher for ESD group than non-ESD group among lesions less than 19 mm (7.4% vs 1.9%, p<0.0001), but not among lesions larger than 20 mm (6.1% vs 7.1%, p=0.6432). The local recurrence rate was significantly lower in ESD group than non-ESD group (p<0.001). Furthermore, for lesions larger than 30 mm, the cumulative local recurrence rate at 2 years was 22.6% in non-ESD group compared to only 1.6% in ESD group (p<0.0001). Conclusions: ER outcomes for SDETs were generally acceptable. ESD by highly experienced endoscopists might be an option for very large SDETs.