Application of artificial intelligence in endoscopic image analysis for the diagnosis of a gastric cancer pathogen-Helicobacter pylori infection.

Helicobacter pylori (H. pylori) infection is the principal cause of chronic gastritis, gastric ulcers, duodenal ulcers, and gastric cancer. In clinical practice, diagnosis of H. pylori infection by a gastroenterologists' impression of endoscopic images is inaccurate and cannot be used for the management of gastrointestinal diseases. The aim of this study was to develop an artificial intelligence classification system for the diagnosis of H. pylori infection by pre-processing endoscopic images and machine learning methods. Endoscopic images of the gastric body and antrum from 302 patients receiving endoscopy with confirmation of H. pylori status by a rapid urease test at An Nan Hospital were obtained for the derivation and validation of an artificial intelligence classification system. The H. pylori status was interpreted as positive or negative by Convolutional Neural Network (CNN) and Concurrent Spatial and Channel Squeeze and Excitation (scSE) network, combined with different classification models for deep learning of gastric images. The comprehensive assessment for H. pylori status by scSE-CatBoost classification models for both body and antrum images from same patients achieved an accuracy of 0.90, sensitivity of 1.00, specificity of 0.81, positive predictive value of 0.82, negative predicted value of 1.00, and area under the curve of 0.88. The data suggest that an artificial intelligence classification model using scSE-CatBoost deep learning for gastric endoscopic images can distinguish H. pylori status with good performance and is useful for the survey or diagnosis of H. pylori infection in clinical practice.

Comparison of the Efficacies of Direct-Acting Antiviral Treatment for HCV Infection in People Who Inject Drugs and Non-Drug Users.

Background and Objectives: Hepatitis C virus (HCV) is a major cause of liver disease worldwide. People who inject drugs (PWIDs) constitute the majority of patients with HCV infection in the United States and Central Asia. There are several obstacles to treating HCV infection in PWIDs because PWIDs are often accompanied by concurrent infection, low compliance, substance abuse, and risky behavior. The aim of the study is to compare the efficacies of direct-acting antiviral (DAA) therapy for HCV infection in PWIDs and those without opioid injection. Materials and Methods: In this retrospective cohort study, we included 53 PWIDs with HCV infections treated on site in a methadone program and 106 age- and sex-matched patients with HCV infections who had no history of opioid injection (ratio of 1:2). All eligible subjects received anti-HCV treatment by DAA agents in our hospital from March 2018 to December 2020. The charts of these patients were carefully reviewed for demographic data, types of DAA agents, and treatment outcomes. The primary outcome measure was sustained virological response (SVR). Results: PWIDs and non-drug users had different HCV genotype profiles (p = 0.013). The former had higher proportions of genotype 3 (18.9% vs. 7.5%) and genotype 6 (24.5% vs. 14.2%) than the latter. The two patient groups had comparable rates of complete drug refilling (100.0% vs. 91.1%) and frequency of loss to follow-up (3.8% vs. 0.9%). However, PWIDs had a lower SVR rate of DAA treatment than non-drug users (92.2% vs. 99.0%; p = 0.04). Further analysis showed that both human immunodeficiency virus (HIV) coinfection and history of PWID were risk factors associated with treatment failure. The subjects with coinfection with HIV had lower SVR rates than those without HIV infection (50.0% vs. 96.5%; p = 0.021). Conclusions: PWIDs with HCV infections have higher proportions of HCV genotype 3 and genotype 6 than non-drug users with infections. DAA therapy can achieve a high cure rate (>90%) for HCV infection in PWID, but its efficacy in PWID is lower than that in non-drug users.