Background: Intrapapillary capillary loops (IPCLs) represent an endoscopically visible feature of early squamous cell neoplasia (ESCN) which correlate with invasion depth-an important factor in the success of curative endoscopic therapy. IPCLs visualised on magnification endoscopy with Narrow Band Imaging (ME-NBI) can be used to train convolutional neural networks (CNNs) to detect the presence and classify staging of ESCN lesions. Methods: A total of 7046 sequential high-definition ME-NBI images from 17 patients (10 ESCN, 7 normal) were used to train a CNN. IPCL patterns were classified by three expert endoscopists according to the Japanese Endoscopic Society classification. Normal IPCLs were defined as type A, abnormal as B1-3. Matched histology was obtained for all imaged areas. Results: This CNN differentiates abnormal from normal IPCL patterns with 93.7% accuracy (86.2% to 98.3%) and sensitivity and specificity for classifying abnormal IPCL patterns of 89.3% (78.1% to 100%) and 98% (92% to 99.7%), respectively. Our CNN operates in real time with diagnostic prediction times between 26.17 ms and 37.48 ms. Conclusion: Our novel and proof-of-concept application of computer-aided endoscopic diagnosis shows that a CNN can accurately classify IPCL patterns as normal or abnormal. This system could be used as an in vivo, real-time clinical decision support tool for endoscopists assessing and directing local therapy of ESCN.

Background and Aims The Seattle protocol for endoscopic Barrett's esophagus surveillance samples a small proportion of the mucosal surface area-risking a potentially high miss rate of early neoplastic lesions. We assess if the new iScan Optical Enhancement system (OE, Pentax) improves the detection of early BE associated neoplasia compared with high definition white light endoscopy (HD-WLE) in both expert and trainee endoscopists to target sampling of suspicious areas. Such a system may both improve early neoplasia detection and reduce the need for random biopsies. Methods 41 patients undergoing endoscopic BE surveillance from Jan 2016-Nov 2017 were recruited from 3 international referral centers. Matched still images in both HD-WLE (n=130) and iScan OE (n=132) were obtained from endoscopic examinations. Two experts, unblinded to the videos and histology, delineated known neoplasia, forming a consensus criterion standard. 7 expert and 7 trainee endoscopists marked one position per image where they would expect a target biopsy to identify dysplastic tissue. The same expert panel then reviewed magnification images and using a previously validated classification system attempted to classify mucosa as dysplastic or non-dysplastic based on the mucosal and vascular patterns observed on magnification endoscopy. Diagnostic accuracy, sensitivity, specificity, NPV, and PPV were calculated. Improvements in dysplasia detection in HD-WLE vs OE and interobserver agreement (IA) were assessed by multilevel logistic regression analysis and Krippendorff's alpha, respectively. Improvements in diagnostic performance were expressed as an odds ratio between the odds of an improvement in OE, compared with the odds of an improvement in WLE Results

Barrett’s esophagus (BE) is the only known precursor to esophageal adenocarcinoma (EAC), whose incidence has increased sharply in the last 4 decades. The annual conversion rate of BE to cancer is significant, but small. The identification of patients at a higher risk of cancer therefore poses a clinical conundrum. Currently, endoscopic surveillance is recommended in BE patients, with the aim of diagnosing either dysplasia or cancer at early stages, both of which are curable with minimally invasive endoscopic techniques. There is a large variation in clinical practice for endoscopic surveillance, and dysplasia as a marker of increased risk is affected by sampling error and high interobserver variability. Screening programs have not yet been formally accepted, mainly due to the economic burden that would be generated by upper gastrointestinal endoscopy. Screening programs have not yet been formally accepted, mainly due to the economic burden that would be generated by widespread indication to upper gastrointestinal endoscopy. In fact, it is currently difficult to formulate an accurate algorithm to confidently target the population at risk, based on the known clinical risk factors for BE and EAC. This review will focus on the clinical and molecular factors that are involved in the development of BE and its conversion to cancer and on how increased knowledge in these areas can improve the clinical management of the disease.

Acute upper gastrointestinal bleeding (AUGIB) is one of the most common medical emergencies in the UK. Despite advancement in technology the management of AUGIB remains a challenge. The clinical community recognise the need for improvement in the treatment of these patients. AUGIB has a significant impact on resources. Endoscopic therapy is the gold standard treatment. The mortality in AUGIB is rarely related to the presenting bleed but significantly associated with concurrent comorbidities. The cost of blood transfusion in the management of patients with AUGIB is significant and misuse of blood products has been documented nationally. Risk stratification tools such as Glasgow-Blatchford Score, Rockall Score and the AIMS65 score have allowed clinicians to triage patients appropriately in order to deliver endoscopic therapy within a suitable time frame. Endoscopic therapeutic modalities such as epinephrine injection, heat thermocoagulation and mechanical clips have had a positive impact on patient’s management. However, in order to continue to improve patient’s outcomes, further developments are needed.