A deep learning system for detection of early Barrett's neoplasia: a model development and validation study.

BACKGROUND: Computer-aided detection (CADe) systems could assist endoscopists in detecting early neoplasia in Barrett's oesophagus, which could be difficult to detect in endoscopic images. The aim of this study was to develop, test, and benchmark a CADe system for early neoplasia in Barrett's oesophagus. METHODS: The CADe system was first pretrained with ImageNet followed by domain-specific pretraining with GastroNet. We trained the CADe system on a dataset of 14 046 images (2506 patients) of confirmed Barrett's oesophagus neoplasia and non-dysplastic Barrett's oesophagus from 15 centres. Neoplasia was delineated by 14 Barrett's oesophagus experts for all datasets. We tested the performance of the CADe system on two independent test sets. The all-comers test set comprised 327 (73 patients) non-dysplastic Barrett's oesophagus images, 82 (46 patients) neoplastic images, 180 (66 of the same patients) non-dysplastic Barrett's oesophagus videos, and 71 (45 of the same patients) neoplastic videos. The benchmarking test set comprised 100 (50 patients) neoplastic images, 300 (125 patients) non-dysplastic images, 47 (47 of the same patients) neoplastic videos, and 141 (82 of the same patients) non-dysplastic videos, and was enriched with subtle neoplasia cases. The benchmarking test set was evaluated by 112 endoscopists from six countries (first without CADe and, after 6 weeks, with CADe) and by 28 external international Barrett's oesophagus experts. The primary outcome was the sensitivity of Barrett's neoplasia detection by general endoscopists without CADe assistance versus with CADe assistance on the benchmarking test set. We compared sensitivity using a mixed-effects logistic regression model with conditional odds ratios (ORs; likelihood profile 95% CIs). FINDINGS: Sensitivity for neoplasia detection among endoscopists increased from 74% to 88% with CADe assistance (OR 2·04; 95% CI 1·73-2·42; p<0·0001 for images and from 67% to 79% [2·35; 1·90-2·94; p<0·0001] for video) without compromising specificity (from 89% to 90% [1·07; 0·96-1·19; p=0·20] for images and from 96% to 94% [0·94; 0·79-1·11; ] for video; p=0·46). In the all-comers test set, CADe detected neoplastic lesions in 95% (88-98) of images and 97% (90-99) of videos. In the benchmarking test set, the CADe system was superior to endoscopists in detecting neoplasia (90% vs 74% [OR 3·75; 95% CI 1·93-8·05; p=0·0002] for images and 91% vs 67% [11·68; 3·85-47·53; p<0·0001] for video) and non-inferior to Barrett's oesophagus experts (90% vs 87% [OR 1·74; 95% CI 0·83-3·65] for images and 91% vs 86% [2·94; 0·99-11·40] for video). INTERPRETATION: CADe outperformed endoscopists in detecting Barrett's oesophagus neoplasia and, when used as an assistive tool, it improved their detection rate. CADe detected virtually all neoplasia in a test set of consecutive cases. FUNDING: Olympus.

Clinical implantation of 92 VACStents in the upper gastrointestinal tract of 50 patients-applicability and safety analysis of an innovative endoscopic concept.

Introduction: Endoscopic vacuum therapy (EVT) has emerged as a promising treatment option for upper gastrointestinal wall defects, offering benefits such as evacuation of secretions and removal of wound debris by suction, and reduction and healing of wound cavities to improve clinical outcomes. In contrast, covered stents have a high rate of migration and lack functional drainage, while endoluminal EVT devices obstruct the GI tract. The VACStent is a novel device that combines the benefits of EVT and stent placement. Its design features a fully covered Nitinol-stent within a polyurethane sponge cylinder, enabling EVT while maintaining stent patency. Methods: This study analyzes the pooled data from three different prospective study cohorts to assess the safe practicality of VACStent placement, complete leak coverage, and effective suction-treatment of esophageal leaks. By pooling the data, the study aims to provide a broader base for analysis. Results: In total, trans-nasal derivation of the catheter, suction and drainage of secretion via vacuum pump were performed without any adversity. In the pooled study cohort of 92 VACStent applications, the mean stent indwelling time was 5.2 days (range 2-8 days) without any dislocation of the device. Removal of the VACStent was done without complication, in one case the sponge was lost but subsequently fully preserved. Minor local erosions and bleeding and one subsequent hemostasis were recorded unfrequently during withdrawal of the device (5.4%, 5/92) but no perforation or pressure ulcer. Despite a high heterogeneity regarding primary disease and pretreatments a cure rate of 76% (38/50 patients) could be achieved. Discussion: In summary, insertion and release procedure was regarded as easy and simple with a low potential of dislocation. The VACStent was well tolerated by the patient while keeping the drainage function of the sponge achieving directly a wound closure by continuous suction and improving the healing process. The implantation of the VACStent provides a promising new procedure for improved clinical treatment in various indications of the upper gastrointestinal wall, which should be validated in larger clinical studies.Clinical Trial Registration: Identifier [DRKS00016048 and NCT04884334].