-
Journal of Healthcare Engineering 2023This study is to evaluate the feasibility of deep learning (DL) models in the multiclassification of reflux esophagitis (RE) endoscopic images, according to the Los...
This study is to evaluate the feasibility of deep learning (DL) models in the multiclassification of reflux esophagitis (RE) endoscopic images, according to the Los Angeles (LA) classification for the first time. The images were divided into three groups, namely, normal, LA classification A + B, and LA C + D. The images from the HyperKvasir dataset and Suzhou hospital were divided into the training and validation datasets as a ratio of 4 : 1, while the images from Jintan hospital were the independent test set. The CNNs- or Transformer-architectures models (MobileNet, ResNet, Xception, EfficientNet, ViT, and ConvMixer) were transfer learning via Keras. The visualization of the models was proposed using Gradient-weighted Class Activation Mapping (Grad-CAM). Both in the validation set and the test set, the EfficientNet model showed the best performance as follows: accuracy (0.962 and 0.957), recall for LA A + B (0.970 and 0.925) and LA C + D (0.922 and 0.930), Marco-recall (0.946 and 0.928), Matthew's correlation coefficient (0.936 and 0.884), and Cohen's kappa (0.910 and 0.850), which was better than the other models and the endoscopists. According to the EfficientNet model, the Grad-CAM was plotted and highlighted the target lesions on the original images. This study developed a series of DL-based computer vision models with the interpretable Grad-CAM to evaluate the feasibility in the multiclassification of RE endoscopic images. It firstly suggests that DL-based classifiers show promise in the endoscopic diagnosis of esophagitis.
Topics: Humans; Esophagitis, Peptic; Deep Learning; Los Angeles; Electric Power Supplies; Glycyrrhetinic Acid
PubMed: 36852218
DOI: 10.1155/2023/7023731 -
BMC Gastroenterology Aug 2020Reflux esophagitis (RE) and absence of Helicobacter pylori (non-H. pylori) are considered to be associated with the progression to long-segment Barrett's esophagus...
Relationship between time-varying status of reflux esophagitis and Helicobacter pylori and progression to long-segment Barrett's esophagus: time-dependent Cox proportional-hazards analysis.
BACKGROUND
Reflux esophagitis (RE) and absence of Helicobacter pylori (non-H. pylori) are considered to be associated with the progression to long-segment Barrett's esophagus (LSBE). However, it is difficult to assess this association because RE and H. pylori status can change during follow-up. Additionally, the association between H. pylori eradication and LSBE remains unclear.
METHODS
A total of 11,493 asymptomatic Japanese subjects who underwent medical check-ups and were endoscopically diagnosed with short-segment Barrett's esophagus (SSBE) between May 2006 and December 2015 were enrolled. The hazards of progression to LSBE were compared between time-varying RE and H. pylori infection/eradication by time-dependent multivariable Cox proportional hazards models.
RESULTS
A total of 7637 subjects who underwent additional medical check-ups after being diagnosed with endoscopic SSBE were analyzed. Subjects with RE and without current/past H. pylori infection were strongly associated with a higher rate of progression to LSBE (adjusted hazard ratio [HR]: 7.17, 95% confidence interval [CI]: 2.48-20.73, p < 0.001 for RE and non-H. pylori vs. non-RE and H. pylori groups). Subjects with H. pylori had a lower rate of progression to LSBE (adjusted HR: 0.48, 95% CI: 0.22-1.07, p = 0.07 for H. pylori vs. non-H. pylori). Hazards of progression to LSBE were still lower in the H. pylori eradication group than that of the non-H. pylori group (adjusted HR: 0.51, 95% CI: 0.18-1.46, p = 0.21).
CONCLUSIONS
RE and non-H. pylori were associated with the progression to LSBE, considering the changes in exposures. H. pylori infection was associated with the prevention of the development of LSBE irrespective of RE. The environment preventive of the development of LSBE persists for at least a few years after H. pylori eradication.
Topics: Barrett Esophagus; Endoscopy; Esophagitis, Peptic; Helicobacter Infections; Helicobacter pylori; Humans
PubMed: 32799812
DOI: 10.1186/s12876-020-01418-5 -
World Journal of Gastroenterology Jan 2017To assess the clinical characteristics of patients with complicated erosive esophagitis (EE) and their associated factors. (Observational Study)
Observational Study
AIM
To assess the clinical characteristics of patients with complicated erosive esophagitis (EE) and their associated factors.
METHODS
This prospective, cross-sectional study included patients diagnosed with EE by upper gastrointestinal endoscopy between October 2014 and March 2015 at 106 Japanese hospitals. Data on medical history, general condition, gastrointestinal symptoms, lifestyle habits, comorbidities, and endoscopic findings were collected using a standard form to create a dedicated database. Logistic regression analysis was used to calculate adjusted odds ratios (aOR) and 95%CI for the association with complicated EE.
RESULTS
During the study period, 1749 patients diagnosed with EE, 38.3% of whom were prescribed proton pump inhibitors (PPIs) were included. Of them, 143 (8.2%) had EE complications. Esophageal bleeding occurred in 84 (4.8%) patients, esophageal strictures in 45 (2.6%) patients, and 14 (0.8%) patients experienced both. Multivariate analysis showed that increased age (aOR: 1.05; 95%CI: 1.03-1.08), concomitant use of psychotropic agents (aOR: 6.51; 95%CI: 3.01-13.61), and Los Angeles grades B (aOR: 2.69; 95%CI: 1.48-4.96), C (aOR: 15.38; 95%CI: 8.62-28.37), and D (aOR: 71.49; 95%CI: 37.47-142.01) were significantly associated with complications, whereas alcohol consumption 2-4 d/wk was negatively associated (aOR: 0.23; 95%CI: 0.06-0.61). Analyzing associated factors with each EE complication separately showed esophageal ulcer bleeding were associated with increased age (aOR: 1.05; 95%CI: 1.02-1.07) and Los Angeles grades B (aOR: 3.60; 95%CI: 1.52-8.50), C (aOR: 27.61; 95%CI: 12.34-61.80), and D (aOR: 119.09; 95%CI: 51.15-277.29), while esophageal strictures were associated with increased age (aOR: 1.07; 95%CI: 1.04-1.10), gastroesophageal reflux symptom (aOR: 2.51; 95%CI: 1.39-4.51), concomitant use of psychotropic agents (aOR: 11.79; 95%CI: 5.06-27.48), Los Angeles grades C (aOR: 7.35; 95%CI: 3.32-16.25), and D (aOR: 20.34; 95%CI: 8.36-49.53) and long-segment Barrett's esophagus (aOR: 4.63; 95%CI: 1.64-13.05).
CONCLUSION
Aging and severe EE were common associated factors, although there were more associated factors in esophageal strictures than esophageal ulcer bleeding. Despite the availability and widespread use of PPIs, EE complications are likely to remain a problem in Japan owing to the aging population and high-stress society.
Topics: Age Factors; Aged; Aged, 80 and over; Alcohol Drinking; Barrett Esophagus; Cross-Sectional Studies; Esophageal Diseases; Esophageal Stenosis; Esophagitis, Peptic; Esophagoscopy; Female; Gastroesophageal Reflux; Gastrointestinal Hemorrhage; Humans; Japan; Male; Middle Aged; Peptic Ulcer; Prospective Studies; Proton Pump Inhibitors; Psychotropic Drugs; Risk Factors; Stress, Psychological
PubMed: 28127205
DOI: 10.3748/wjg.v23.i2.318 -
The Korean Journal of Gastroenterology... May 2011
Topics: Alcohol Drinking; Antacids; Endoscopy, Gastrointestinal; Esophagitis, Peptic; Gastroesophageal Reflux; Humans; Risk Factors; Sex Factors
PubMed: 21623134
DOI: 10.4166/kjg.2011.57.5.269 -
BMC Gastroenterology Feb 2019The prevalence rate of reflux esophagitis (RE) in Asia, including Taiwan, has increased dramatically in recent years. However, few studies have discussed on its...
BACKGROUND
The prevalence rate of reflux esophagitis (RE) in Asia, including Taiwan, has increased dramatically in recent years. However, few studies have discussed on its relationship with metabolic syndrome (MetS). This study aimed to evaluate the correlation between RE and MetS and its components.
METHODS
We conducted a cross-sectional study during 2013 to 2014 in Taiwan. A total of 4895 subjects who completed upper gastrointestinal endoscopy at the Health Examination Center of Changhua Christian Hospital were enrolled. RE was defined according to the upper gastrointestinal endoscopic findings and MetS was defined according to the Taiwanese criteria. Univariate and multivariate logistic regression analyses were applied to calculate odds ratios and 95% confidence intervals for each variable to assess the associated features for RE. We analyzed the relationship between the number of MetS components and the severity of RE using the chi-square test for trend.
RESULTS
The prevalence rates of MetS and RE were respectively 28.5 and 59.6%. According to univariate logistic regression analysis, MetS was significantly associated with RE and remained a positive association in multivariate logistic regression analysis (adjusted OR = 1.251; 95% CI = 1.071-1.462; p = 0.005). Furthermore, among the five MetS components, elevated blood pressure (adjusted OR = 1.163; 95% CI = 1.023-1.323; p = 0.021), abdominal obesity (adjusted OR = 1.173; 95% CI = 1.020-1.349; p = 0.026) and hyperglycemia (adjusted OR = 1.306; 95% CI = 1.142-1.495; p < 0.001) were positively associated with the presence of RE. A weak association was also found between elevated triglycerides and RE after adjusting for age and gender (adjusted OR = 1.171; 95% CI = 1.022-1.343; p = 0.023). Reduced high-density lipoprotein cholesterol showed no significant difference between groups with and without RE. Older age (≥65 years), male gender, higher body mass index, higher uric acid, smoking, alcohol drinking, and hiatal hernia were found to be significant associated factors for RE. In addition, a dose-response relation between the number of MetS components and the presence of RE was demonstrated in the multivariate analysis. Furthermore, we performed a trend analysis and found the severity of RE got worse as the number of MetS components increased (p < 0.001).
CONCLUSION
This study suggests that MetS is significantly related to the presence and the severity of RE.
Topics: Adult; Aged; Comorbidity; Cross-Sectional Studies; Esophagitis, Peptic; Female; Humans; Hyperglycemia; Hypertension; Male; Metabolic Syndrome; Middle Aged; Obesity, Abdominal; Prevalence; Retrospective Studies; Severity of Illness Index; Taiwan
PubMed: 30782138
DOI: 10.1186/s12876-019-0950-z -
Postgraduate Medical Journal Apr 1974The endoscopic diagnosis of hiatal hernia (sliding type) relies on direct and indirect criteria during the oesophagoscopic or gastroscopic approach. A wide separation...
The endoscopic diagnosis of hiatal hernia (sliding type) relies on direct and indirect criteria during the oesophagoscopic or gastroscopic approach. A wide separation between the anatomic and mucosal oesophago-gastric junction, the presence of a so-called Schatzki-ring, a wandering junction with sliding gastric mucosal folds during inspiration and two ring-like structures are important. Whereas radiology seems to be superior to endoscopy in the diagnosis of asymptomatic hiatus hernia, oesophagoscopy may reveal sequelae of reflux like oesophagitis, erosions, ulcers and strictures. Peptic oesophagitis is found in about 10% to cause upper gastrointestinal haemorrhage.
Topics: Candidiasis; Esophagitis, Peptic; Esophagogastric Junction; Esophagoscopy; Gastroesophageal Reflux; Hernia, Diaphragmatic; Hernia, Hiatal; Humans; Middle Aged; Radiography
PubMed: 4449767
DOI: 10.1136/pgmj.50.582.199 -
Seminars in Roentgenology Oct 1994
Review
Topics: Barrett Esophagus; Esophagitis, Peptic; Esophagus; Humans; Radiography
PubMed: 7809647
DOI: 10.1016/s0037-198x(05)80047-1 -
Digestive Diseases (Basel, Switzerland) 2021Refractory reflux esophagitis (RRE), unresponsive to conventional proton-pump inhibitors (PPIs), is a complication in esophagectomy with gastric pull-up. Vonoprazan...
BACKGROUND
Refractory reflux esophagitis (RRE), unresponsive to conventional proton-pump inhibitors (PPIs), is a complication in esophagectomy with gastric pull-up. Vonoprazan (VPZ), a novel potassium-competitive acid blocker, has been available in Japan since 2015. Here, we investigated the efficacy of VPZ on PPI-resistant RRE after esophagectomy with gastric pull-up.
METHODS
This was a single-center retrospective study. We used the revised Los Angeles (r-LA) classification based on the Los Angeles classification and the modified Los Angeles classification to evaluate abnormal forms of mucosal breaks such as lateral spreading consistently. Patients who underwent esophagectomy with gastric pull-up and had RRE grade B-D as per the r-LA classification, despite using standard-dose PPIs or double dose of rabeprazole, were included. Sixteen patients who switched to VPZ (20 mg/day) and 14 patients who continued PPIs were assigned to the VPZ and PPI groups, respectively. Endoscopic observations were reviewed by 3 endoscopists using the r-LA classification to ensure consistent diagnosis, while the treatment arm and patient information were blinded to evaluators. We defined mucosal breaks that improved by at least one grade after treatment as improved mucosa and recovery to grade M or N as mucosal healing.
RESULTS
The percentage of patients with improved mucosa in the VPZ and PPI groups was 81.3 and 14.3%, respectively (p < 0.001). The rate of mucosal healing was 68.8 and 7.1%, respectively (p = 0.001).
CONCLUSION
VPZ significantly improved PPI-resistant RRE after esophagectomy with gastric pull-up.
Topics: Esophagectomy; Esophagitis, Peptic; Humans; Proton Pump Inhibitors; Pyrroles; Retrospective Studies; Sulfonamides; Treatment Outcome
PubMed: 33567428
DOI: 10.1159/000515146 -
PloS One 2013Despite the high prevalence of gastroesophageal reflux disease (GERD), its risk factors are still a subject of controversy. This is probably due to inadequate...
BACKGROUND
Despite the high prevalence of gastroesophageal reflux disease (GERD), its risk factors are still a subject of controversy. This is probably due to inadequate distinction between reflux esophagitis (RE) and non-erosive reflux disease (NERD), and is also due to inadequate evaluation of adjacent stomach. Our aim is therefore to define background factors of RE and NERD independently, based on the evaluation of Helicobacter pylori infection and gastric atrophy.
METHODS
We analyzed 10,837 healthy Japanese subjects (6,332 men and 4,505 women, aged 20-87 years) who underwent upper gastrointestinal endoscopy. RE was diagnosed as the presence of mucosal break, and NERD was diagnosed as the presence of heartburn and/or acid regurgitation in RE-free subjects. Using GERD-free subjects as control, background factors for RE and NERD were separately analyzed using logistic regression to evaluate standardized coefficients (SC), odds ratio (OR), and p-value.
RESULTS
Of the 10,837 study subjects, we diagnosed 733 (6.8%) as RE and 1,722 (15.9%) as NERD. For RE, male gender (SC = 0.557, OR = 1.75), HP non-infection (SC = 0.552, OR = 1.74), higher pepsinogen I/II ratio (SC = 0.496, OR = 1.64), higher BMI (SC = 0.464, OR = 1.60), alcohol drinking (SC = 0.161, OR = 1.17), older age (SC = 0.148, OR = 1.16), and smoking (SC = 0.129, OR = 1.14) are positively correlated factors. For NERD, HP infection (SC = 0.106, OR = 1.11), female gender (SC = 0.099, OR = 1.10), younger age (SC = 0.099, OR = 1.10), higher pepsinogen I/II ratio (SC = 0.099, OR = 1.10), smoking (SC = 0.080, OR = 1.08), higher BMI (SC = 0.078, OR = 1.08), and alcohol drinking (SC = 0.076, OR = 1.08) are positively correlated factors. Prevalence of RE in subjects with chronic HP infection and successful HP eradication denotes significant difference (2.3% and 8.8%; p<0.0001), whereas that of NERD shows no difference (18.2% and 20.8%; p = 0.064).
CONCLUSIONS
Significantly associated factors of NERD are considerably different from those of RE, indicating that these two disorders are pathophysiologically distinct. Eradication of Helicobacter pylori may have disadvantageous effects on RE but not on NERD.
Topics: Adult; Aged; Cross-Sectional Studies; Esophagitis, Peptic; Female; Gastroesophageal Reflux; Helicobacter Infections; Helicobacter pylori; Humans; Japan; Male; Middle Aged; Prevalence; Risk Factors
PubMed: 23922844
DOI: 10.1371/journal.pone.0069891 -
Digestive Diseases and Sciences Nov 2016Predictors of erosive esophagitis (EE) and Barrett's esophagus (BE) and the influence of number of risk factors in the community are not well defined.
BACKGROUND
Predictors of erosive esophagitis (EE) and Barrett's esophagus (BE) and the influence of number of risk factors in the community are not well defined.
METHODS
Rates of BE and EE among community residents identified in a randomized screening trial were defined. The risk of EE and BE associated with single and multiple risk factors (gender, age, GERD, Caucasian ethnicity, ever tobacco use, excess alcohol use, family history of BE or EAC, and central obesity) was analyzed.
RESULTS
Sixty-eight (33 %) of 205 subjects had EE and/or BE. BE prevalence was 7.8 % with dysplasia present in 1.5 %. Rates were comparable between subjects with and without GERD. Male sex and central obesity were independent risk factors. The odds of EE or BE were 3.7 times higher in subjects with three or four risk factors and 5.7 times higher in subjects with five or more risk factors compared with those with two or less factors.
CONCLUSIONS
EE and BE are prevalent in the community regardless of the presence of GERD. Risk appeared to be additive, increasing substantially with three or more risk factors.
Topics: Age Factors; Aged; Aged, 80 and over; Alcohol Drinking; Barrett Esophagus; Esophagitis, Peptic; Ethnicity; Female; Gastroesophageal Reflux; Humans; Independent Living; Logistic Models; Male; Middle Aged; Multivariate Analysis; Obesity, Abdominal; Odds Ratio; Prevalence; Risk Factors; Sex Factors; Smoking; White People
PubMed: 27510751
DOI: 10.1007/s10620-016-4266-3