Colorectal cancer (CRC) prevention depends on effective colonoscopy; yet variability in adenoma detection rates (ADRs) and missed lesions remain significant hurdles. Artificial intelligence-powered computer-aided detection (CADe) systems offer promising advancements in enhancing polyp detection. This review examines the role of CADe in improving ADR and reducing adenoma miss rates (AMRs) while addressing its broader clinical implications. CADe has demonstrated consistent improvements in ADRs and AMRs; largely by detecting diminutive polyps, but shows limited efficacy in identifying advanced adenomas or sessile serrated lesions. Challenges such as operator deskilling and the need for enhanced algorithms persist. Combining CADe with adjunctive techniques has shown potential for further optimizing performance. While CADe has standardized detection quality; its long-term impact on CRC incidence and mortality remains inconclusive. Future research should focus on refining CADe technology and assessing its effectiveness in reducing the global burden of CRC.

The European Society of Gastrointestinal Endoscopy (ESGE) has identified a critical need to establish structured training for safe and effective use of artificial intelligence (AI) in endoscopy. This manuscript presents the results of a formal Delphi consensus process and outlines the official ESGE position, offering a comprehensive curriculum for acquiring and maintaining the competence needed to exploit the benefit of using AI tools in endoscopy. The proposed framework defines the prerequisites in the preadoption phase, core training components, and requirements to maintain optimal implementation. Key recommendations include: (1) ensuring basic competency in standard endoscopy procedures; (2) acquiring foundational knowledge of AI principles; (3) implementing educational programs to enhance AI literacy; (4) recognizing and mitigating cognitive biases in human–AI interaction; (5) avoiding over-reliance on AI in clinical decision-making; and (6) continuous monitoring of key performance indicators throughout AI system integration.

Background This study assessed the effectiveness of large language models (LLMs) in generating lay summaries for patient education on the management of precancerous lesions and early neoplasia in the stomach. Methods In this pilot study, we used a two-period, crossover, blinded design to compare a ChatGPT-4o summary versus a Digestive Cancers Europe (DiCE) summary. Two panels rated the materials: expert physicians and DiCE Patient Advisory Committee members. Experts scored accuracy, completeness, comprehensibility, and satisfaction (across five sections); patients rated overall completeness, comprehensibility, and satisfaction. Paired comparisons used mixed-effects estimates. Readability was assessed with Flesch–Kincaid grade level (FKGL) and SMOG index. Results Median expert ratings were similar between materials across metrics. For the overall summary, median (range; IQR) scores were: accuracy 5 (4–6; 1) for ChatGPT-4o vs. 5 (3–6; 1) for DiCE (P = 0.10); completeness 4 (3–5; 1) vs. 4 (2–5; 1; P = 0.27); comprehensibility 4 (3–5; 1) vs. 4 (2–5; 1; P = 0.33); and satisfaction 4 (2–5; 1) vs. 3 (1–5; 2; P = 0.53). Patient ratings mirrored experts, with very similar results. Readability failed to meet guideline recommendations for both summaries on both FKGL and SMOG scores. Conclusion ChatGPT-4o produced patient materials comparable to DiCE, but both require readability optimization; a human-in-the-loop workflow and future tests across prompts and models are warranted. ^‡ joint first authors. ^* joint senior authors. Received: 14 March 2025 Accepted after revision: 29 December 2025 Article published online: 13 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Biliary drainage in patients with distal malignant biliary obstruction (DMBO) carries a higher risk of difficult biliary cannulation (DBC) during endoscopic retrograde cholangiopancreatography (ERCP). After the failure of standard cannulation, endoscopists may proceed with advanced cannulation techniques and/or with endoscopic ultrasound-guided biliary drainage (EUS-BD).This was a retrospective study of consecutive patients with DMBO and a dilated common bile duct (CBD; >12 mm) who underwent ERCP for endoscopic biliary drainage in four European centers. The rates of DBC, technical and clinical success, and procedure-related adverse events (AEs) were assessed. The predictive factors for AEs were also investigated through regression analysis. The EUS-BD approach was considered either as the first option after standard cannulation failure or as the final option after advanced cannulation failure.1016 patients with DMBO were included in the study, with 524 (51.6%) matching the definition of DBC. Clinical success was achieved in 956 patients (94.1%). Procedure-related AEs were experienced by 167 patients (16.4%). Patients with DBC had a higher risk of AEs (P=0.003); however, patients undergoing “early” EUS-BD showed a risk of AEs comparable with those managed with standard cannulation (P=0.38). An attempt at any advanced cannulation technique was independently associated with the occurrence of AEs (P=0.001).The risk of AEs is higher in patients with DMBO and DBC, this appears to be mainly related to the advanced cannulation techniques. In patients with a dilated CBD (>12 mm), “early” EUS-BD may minimize the risk of AEs.

Gastrointestinal endoscopy is a major contributor of medical waste and greenhouse gas (GHG) emissions. Sustainable waste management practices could help reduce its environmental footprint.A prospective exploratory study was conducted at IRCCS Humanitas Research Hospital from April to June 2023. The intervention included installing recycling bins for paper and plastic in five endoscopy suites, alongside existing regulated medical waste (RMW) and non-regulated medical waste bins. Comprehensive staff training on waste segregation was provided, and a “Green Endoscopy” nurse monitored compliance during the intervention. Data on RMW, its emissions (in kg carbon dioxide equivalents [CO2e] per procedure), and costs were collected and compared with pre-intervention values.RMW production decreased by 24.1%, from 1.91 kg per procedure (95%CI 1.54–2.28) to 1.45 kg per procedure (95%CI 1.16–1.73; P = 0.02), with sustained reductions during follow-up (1.48 kg per procedure; 95%CI 0.84–2.12). The intervention led to a 33.3% reduction in GHG emissions (with a net difference of 1.23 kg CO2e per procedure; P = 0.02), and cost savings of approximately €3542 (±384) were achieved. This effect was consistent across different procedure types.Targeted interventions such as staff education and structured waste segregation protocols substantially reduced RMW and associated GHG emissions in a high-volume endoscopy unit. These results suggest that simple, low-cost changes may improve environmental sustainability in clinical practice.

Background Lower gastrointestinal bleeding (LGIB) is a common condition linked to increased morbidity, healthcare costs, and mortality. Currently, no prospectively validated prognostic model exists to predict mortality in patients with LGIB. Our aim was to develop and validate a risk score that could accurately predict in-hospital mortality of patients admitted for LGIB. Methods Patient data from a nationwide cohort study in 15 centers in Italy (2019–2020) were used to derive the risk score, the Acute Lower gastrointestinal Bleeding and In-hospital mortality (ALIBI) score; the model was then externally validated in a cohort of consecutive patients hospitalized for LGIB in 12 centers from six countries (Italy, Spain, France, Greece, Iran, and Brazil) from 2022 to 2024. The main outcome was in-hospital mortality; we also reported rebleeding rates and the in-hospital mortality rate stratified by risk score and timing of colonoscopy. Results Among 1198 patients in the derivation cohort, 105 (8.8%) re-bled and 41 (3.4%) died. Age, Charlson Co-morbidity Index, in-hospital onset, hemodynamic instability, and creatinine level were independent predictors of in-hospital mortality. The model demonstrated excellent discrimination (area under the receiver operating curve [AUROC] 0.81, 95%CI 0.75–0.87) and calibration. In the validation cohort (n = 752 patients), the model's good discrimination (AUROC 0.79, 95%CI 0.72–0.86) and calibration were confirmed. Patients were categorized as low (0–4 points; 1% mortality), intermediate (5–9 points; 4.6% mortality), or high risk (10–13 points; 19.1% mortality). Conclusion A new validated score effectively predicts in-hospital mortality in patients with LGIB, aiding in their risk stratification and management.