A convolutional neural network for bleeding detection in capsule endoscopy using real clinical data.

BACKGROUND: The goal of the present study was to develop a convolutional neural network for the detection of bleedings in capsule endoscopy videos using realistic clinical data from one single-centre. METHODS: Capsule endoscopy videos from all 133 patients (79 male, 54 female; meanage = 53.73 years, SDage = 26.13) who underwent capsule endoscopy at our institution between January 2014 and August 2018 were screened for pathology. All videos were screened for pathology by two independent capsule experts and confirmed findings were checked again by a third capsule expert. From these videos, 125 pathological findings (individual episodes of bleeding spanning a total of 5696 images) and 103 non-pathological findings (sections of normal mucosal tissue without pathologies spanning a total of 7420 images) were used to develop and validate a neural network (Inception V3) using transfer learning. RESULTS: The overall accuracy of the model for the detection of bleedings was 90.6% [95%CI: 89.4%-91.7%], with a sensitivity of 89.4% [95%CI: 87.6%-91.2%] and a specificity of 91.7% [95%CI: 90.1%-93.2%]. CONCLUSION: Our results show that neural networks can detect bleedings in capsule endoscopy videos under realistic, clinical conditions with an accuracy of 90.6%, potentially reducing reading time per capsule and helping to improve diagnostic accuracy.

Submucosal Injection of the RNA Oligonucleotide GUT-1 in Active Ulcerative Colitis Patients: A Randomized, Double-Blind, Placebo-Controlled Phase 2a Induction Trial.

BACKGROUND AND AIMS: Carbohydrate sulfotransferase 15 [CHST15] biosynthesizes sulphated matrix glycosaminoglycans and is implicated in intestinal inflammation and fibrosis. Here, we evaluate the efficacy and safety of the double-stranded RNA oligonucleotide GUT-1, a specific blocker of CHST15, as induction therapy in patients with ulcerative colitis [UC]. METHODS: In this randomized, double-blind, placebo-controlled, phase 2a study, we enrolled endoscopically active UC patients, refractory to conventional therapy, in five hospital centres across Germany. Patients were randomized 1:1:1 using a block randomized technique to receive a single dosing of 25 nM GUT-1, 250 nM GUT-1, or placebo by endoscopic submucosal injections. The primary outcome measure was improvement of endoscopic lesions at weeks 2 or 4. The secondary outcome measures included clinical and histological responses. Safety was assessed in all patients who received treatment. RESULTS: Twenty-eight patients were screened, 24 were randomized, and 21 were evaluated. Endoscopic improvement at weeks 2 or 4 was achieved by 71.4% in the GUT-1 250 nM, 0% in the GUT-1 25 nM, and 28.6% in the placebo group. Clinical remission was shown by 57.1% in the GUT-1 250 nM, 0% in the GUT-1 25 nM, and 14.3% in the placebo groups. Histological improvement was shown by 42.9% in the GUT-1 250 nM, 0% in the GUT-1 25 nM, and 0% in the placebo groups. GUT-1 250 nM reduced CHST15 expression significantly and suppressed mucosal inflammation and fibrosis. GUT-1 application was well tolerated. CONCLUSION: Single dosing by submucosal injection of GUT-1 repressed CHST15 mucosal expression and may represent a novel induction therapy by modulating tissue remodelling in UC.

Successful use of TNFα blockade in a severe case of idiopathic non-granulomatous ulcerative jejunoileitis associated with thrombotic thrombocytopenic purpura.

We describe the case of 50-year-old female patient who presented with severe gastrointestinal symptoms and progressive weight loss of unknown origin. Shortly after admission, she developed an acute flare of thrombotic thrombocytopaenic purpura (TTP) that had to be treated by plasma exchange therapy and rituximab administration. While the signs of TTP subsided, the gastrointestinal symptoms worsened with abdominal cramps, massive gastric retention, malnourishment and a stenosis due to extensive inflammation and wall thickening of the small bowel. Extensive diagnostic efforts yielded no specific cause, so the patient-based on the histopathological findings-was diagnosed with idiopathic non-granulomatous ulcerative jejunoileitis. Following a highly complicated clinical course over several months, successful remission of the inflammatory activity and recovery of the patient could be obtained by TNF-alpha blockade.

Effects of Helicobacter pylori Eradication in Chronic Spontaneous Urticaria: Results from a Retrospective Cohort Study.

BACKGROUND AND OBJECTIVE: Helicobacter pylori (Hp) infection has been hypothesised to play a major role in the pathogenesis of chronic spontaneous urticaria (CSU). Despite only weak evidence from Hp eradication studies, screening for Hp infection is still recommended in several CSU guidelines. The aim of this study was to investigate the effect of Hp eradication in combination with standard CSU treatment in Hp-positive compared with Hp-negative patients, applying the latest guidelines for both diseases. METHODS: 138 consecutive patients with CSU were enrolled in this retrospective cohort study. All patients underwent gastroscopy and Hp status was determined by urease testing and histologic examination. Seventy-five patients were diagnosed as Hp negative and 47 patients fulfilled criteria for definite Hp infection, 45 of whom received eradication therapy. Sixteen patients who received eradication therapy without an appropriate indication served as the medication control. All patients received symptomatic treatment with antihistamines and/or glucocorticoids regardless of Hp status. Partial response (PR) was defined as subjective amelioration of CSU symptoms; patients returning for further CSU treatment within 6 months were considered non-responders/relapsers (NRs). RESULTS: The prevalence of Hp infection was comparable with Hp seroprevalence data reported for healthy western populations. Standard treatment of CSU led to relief of symptoms independent of Hp status. Hp eradication by standard triple therapy had no additional effect on PR (p = 0.32) or NR (p = 0.50). CONCLUSIONS: Hp eradication has no discernible effect on CSU beyond that of standard CSU therapy. Therefore, Hp eradication should only be initiated in accordance with currently accepted indications of Hp treatment guidelines.