Search

Background and Aims: The aim of this study was to develop and externally validate a computer-aided detection (CAD) system for the detection and localization of Barrett's neoplasia and assess its performance compared with that of general endoscopists in a statistically powered multicenter study by using real-time video sequences., Methods: In phase 1, the hybrid visual geometry group 16-SegNet model was trained by the use of 75,198 images and videos (96 patients) of neoplastic and 1,014,973 images and videos (65 patients) of nonneoplastic Barrett's esophagus. In phase 2, image-based validation was performed on a separate dataset of 107 images (20 patients) of neoplastic and 364 images (14 patients) of nonneoplastic Barrett's esophagus. In phase 3 (video-based external validation) we designed a real-time video-based study with 32 videos (32 patients) of neoplastic and 43 videos (43 patients) of nonneoplastic Barrett's esophagus from 4 European centers to compare the performance of the CAD model with that of 6 nonexpert endoscopists. The primary endpoint was the sensitivity of CAD diagnosis of Barrett's neoplasia., Results: In phase 2, CAD detected Barrett's neoplasia with sensitivity, specificity, and accuracy of 95.3%, 94.5%, and 94.7%, respectively. In phase 3, the CAD system detected Barrett's neoplasia with sensitivity, specificity, negative predictive value, and accuracy of 93.8%, 90.7%, 95.1%, and 92.0%, respectively, compared with the endoscopists' performance of 63.5%, 77.9%, 74.2%, and 71.8%, respectively (P < .05 in all parameters). The CAD system localized neoplastic lesions with accuracy, mean precision, and mean intersection over union of 100%, 0.62, and 0.54, respectively, when compared with at least 1 of the expert markings. The processing speed of the CAD detection and localization were 5 ms/image and 33 ms/image, respectively., Conclusion: To our knowledge, this is the first study describing external (multicenter) validation of AI algorithms for the detection of Barrett's neoplasia on real-time endoscopic videos. The CAD system in this study significantly outperformed nonexpert endoscopists on real-time video-based assessment, achieving >90% sensitivity for neoplasia detection. This result needs to be validated during real-time endoscopic assessment., (Copyright © 2023 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.)

Background and Aims: In high-risk surgical patients, the treatment of choice of acute cholecystitis is percutaneous transhepatic gallbladder drainage (PTGBD). Recently, a novel endoscopic device containing a lumen-apposing metal stent with an electrocautery (ECE-LAMS) on the tip has been developed., Methods: High-risk surgical patients with acute cholecystitis who underwent EUS-guided gallbladder drainage (EUS-GBD) with the novel device were retrospectively retrieved from 7 tertiary care referral centers. Main endpoints were technical and clinical success rates, rate of procedural adverse events, and short- and long-term adverse events., Results: Seventy-five patients (mean age, 75 ± 11 years; 36 men) underwent EUS-GBD. The procedure was technically and clinically successful in 98.7% and 95.9% of cases, respectively. Three patients without resolution of cholecystitis died, and 2 patients had procedure-related adverse events: 1 perforation requiring surgery and 1 major bleeding resolved conservatively. The mean follow-up for the entire cohort was 201 ± 226 days. Seven patients (9.6%) died within the first 30 days; 50 patients (71.4%) were alive at the last date of follow-up. Short- and long-term adverse events occurred in 6 patients: 3 had recurrent cholecystitis, 2 had migration of the stent, and 1 developed Bouveret syndrome, all managed nonsurgically. Overall, 8 adverse events (10.7%) occurred in the entire cohort of patients., Conclusions: The novel ECE-LAMS for high-risk surgical patients with acute cholecystitis is safe, with a high technical and clinical success rate. Future multicenter studies comparing EUS-GBD versus PTGBD are warranted to determine which procedure is safer and clinically more effective for patients with high surgical risk acute cholecystitis., (Copyright © 2017 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.)

Background and Aims: A lumen-apposing, self-expanding metal stent incorporated in an electrocautery-enhanced delivery system for EUS-guided drainage of pancreatic fluid collections (PFCs) recently has become available. The aim of this study was to analyze the safety and clinical effectiveness of this newly developed device in this clinical setting., Methods: This was a retrospective analysis of all consecutive patients with PFCs who underwent EUS-guided drainage using the study device in 13 European centers., Results: Ninety-three patients with PFCs (80% with complex collections) underwent drainage using the study device. Penetration of the PFC was accomplished directly with the study device in 74.2% of patients, and successful stent placement was accomplished in all but 1 patient, mostly without fluoroscopic assistance. Direct endoscopic necrosectomy (DEN) was carried out in 31 of 52 cases (59.6%) of walled-off necrosis and in 2 of 4 cases (50%) of acute peripancreatic fluid collection. Complete resolution of the PFC was obtained in 86 cases (92.5%), with no recurrence during follow-up. Treatment failure occurred in 6 patients because of persistent infection requiring surgery (n = 3), perforation and massive bleeding caused by the nasocystic drainage catheter (NCDC) (n = 2), and the need for a larger opening to extract large necrotic tissue pieces (n = 1). Major adverse events occurred in 5 patients (perforation and massive bleeding caused by the NCDC in 2 patients, 1 pneumoperitoneum and 1 stent dislodgement during DEN, and 1 postdrainage infection) and were mostly not related to the drainage procedure., Conclusions: EUS-guided drainage with the electrocautery-enhanced delivery system is a safe, easy to perform, and a highly effective minimally invasive treatment modality for PFCs., (Copyright © 2015 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.)

Background: Endosonographic evaluation and sampling of right colonic subepithelial lesions is technically difficult., Objective: To evaluate the feasibility, safety, and tissue yield of a prototype front-view, forward-array, curved linear array echoendoscope in the evaluation and sampling of right colonic subepithelial lesions., Setting: Tertiary referral center., Design: Procedural and outcome data on all patients undergoing EUS evaluation of right-sided colonic and pericolonic lesions were collected during a 1-year study period., Main Outcome Measurements: Patient demographics, clinical indication, EUS findings, EUS-FNA yield, technical success, and procedural complications., Results: A total of 15 patients underwent EUS examination of right-sided colonic lesions with the prototype echoendoscope. The lesions were located in the cecum (n = 12) and the ascending colon (n = 3). The cecum was reached in all examinations within 10 minutes. Twelve patients had subepithelial lesions detected during colonoscopy. Findings included 6 extrinsic compressions from an adjacent normal structure, 1 calcified lymph node, 1 ovarian cyst, 1 prolapsed appendiceal orifice, 1 GI stromal tumor, 1 appendiceal mucocele, and 1 lymphoma. Two patients were evaluated for a pericolonic lesion seen on CT; findings included focal diverticulitis and a metastatic lymph node. In the patient evaluated for an infiltrative mass with previous nondiagnostic biopsies, colonic histoplasmosis was diagnosed. FNA was performed in 6 patients and provided diagnostic samples in 5 (83%). No procedure- or FNA-related complications were reported., Conclusions: Endosonographic evaluation and sampling of right-sided colonic subepithelial lesions with the prototype front-view, forward-array, curved linear array echoendoscope is feasible and safe. The yield of FNA is high, consistent with applications in the upper GI tract., (Copyright 2010 American Society for Gastrointestinal Endoscopy. Published by Mosby, Inc. All rights reserved.)