Consistency between the endoscopic Kyoto classification and pathological updated Sydney system for gastritis: A cross-sectional study

A deep learning approach for gastroscopic manifestation recognition based on Kyoto Gastritis Score

OBJECTIVE

The risk of gastric cancer can be predicted by gastroscopic manifestation recognition and the Kyoto Gastritis Score. This study aims to validate the applicability of AI approaches for recognizing gastroscopic manifestations according to the definition of Kyoto Gastritis Score, with the goal of improving early gastric cancer detection and reducing gastric cancer mortality.

METHODS

In this retrospective study, 29013 gastric endoscopy images were collected and carefully annotated into five categories according to the Kyoto Gastritis Score, i.e. atrophy (A), diffuse redness (DR), enlarged folds (H), intestinal metaplasia (IM), and nodularity (N). As a multi-label recognition task, we propose a deep learning approach composed of five GAM-EfficientNet models, each performing a multiple classification to quantify gastroscopic manifestations, i.e. no presentation or the severity score 0-2. This approach was compared with endoscopists of varying years of experience in terms of accuracy, specificity, precision, recall, and F1 score.

RESULTS

The approach demonstrated good performance in identifying the five manifestations of the Kyoto Gastritis Score, with an average accuracy, specificity, precision, recall, and F1 score of 78.70%, 91.92%, 80.23%, 78.70%, and 0.78, respectively. The average performance of five experienced endoscopists was 72.63%, 90.00%, 77.68%, 72.63%, and 0.73, while that of five less experienced endoscopists was 66.60%, 87.44%, 70.88%, 66.60%, and 0.66, respectively. The sample t-test indicates that the approach's average accuracy, specificity, precision, recall, and F1 score for identifying the five manifestations were significantly higher than those of less experienced endoscopists, experienced endoscopists, and all endoscopists on average (p < 0.05).

CONCLUSION

Our study demonstrates the potential of deep learning approaches on gastric manifestation recognition over junior, even senior endoscopists. Thus, the deep learning approach holds potential as an auxiliary tool, although prospective validation is still needed to assess its clinical applicability.

Incidence of Gastric Adenocarcinoma in Those With Gastric Atrophy: A Systematic Review

Gastric atrophy (GA), or atrophic gastritis, is a pre-neoplastic lesion of gastric cancer (GC). It is part of the Correa cascade, which culminates in intestinal-type gastric adenocarcinoma. The cascade posits that intestinal-type gastric adenocarcinoma develops along a defined pathway of pre-neoplastic stages. The cascade begins with chronic gastritis, most commonly caused by Helicobacter pylori (H. pylori) infection, and proceeds through GA, gastric intestinal metaplasia (GIM), both complete and incomplete, dysplasia, both low and high-grade, and culminating in intestinal-type gastric adenocarcinoma. Attempts in Europe have been made to identify patients at risk of developing GC and target them with surveillance oesophagogastroduodenoscopy (OGD). However, there remains uncertainty about GA's risk of developing into GC. This poses issues in terms of guiding the need for and determining intervals for surveillance OGDs, which are a costly form of surveillance. As such, we attempted to gather all available studies assessing the risk of GC developing from GA, which is the first step in the Correa cascade. This study was a comprehensive systematic review of published papers, reported per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. This systematic review, which included a substantial 25,455 patients across 18 studies, found that the relative risk (RR) of GC in those with GA, using standardised incidence ratios as a measure of RR, was 15.1, with a 95% confidence interval ranging from 13.5 to 16.9. We conclude that GA does increase the risk of developing GC, and this risk may be higher than previously appreciated. Further large-scale studies are needed in Western cohorts of patients to precisely define this risk and guide the need for surveillance programs. These future studies must be standardised to account for H. pylori status, the topographical distribution of the GA, and the methods for assessing the degree of GA.

Characteristic endoscopic findings in Helicobacter pylori diagnosis in clinical practice

INTRODUCTION

Helicobacter pylori is a major risk factor for gastric cancer. In addition to eradication therapy, early-phase detection of gastric cancer through screening programs using high-vision endoscopy is also widely known to reduce mortality. Although European and US guidelines recommend evaluation of atrophy and intestinal metaplasia by high-vision endoscopy and pathological findings, the guideline used in Japan - the Kyoto classification of gastritis - is based on endoscopic evaluation, and recommends the grading of risk factors. This system requires classification into three endoscopic groups: H. pylori-negative, previous H. pylori infection (inactive gastritis), and current H. pylori infection (active gastritis). Major endoscopic findings in active gastritis are diffuse redness, enlarged folds, nodularity, mucosal swelling, and sticky mucus, while those in H pylori-related gastritis - irrespective of active or inactive status - are atrophy, intestinal metaplasia, and xanthoma.

AREAS COVERED

This review describes the endoscopic characteristics of current H. pylori infection, and how characteristic endoscopic findings should be evaluated.

EXPERT OPINION

Although the correct evaluation of endoscopic findings related to H. pylori remains necessary, if findings of possible infection are observed, it is important to diagnose infection by detection methods with high sensitivity and specificity, including the stool antigen test and urea breath test.

Long-term endoscopic gastric mucosal changes up to 20 years after Helicobacter pylori eradication therapy

Helicobacter pylori eradication therapy reduces the risk of gastric cancer. However, it is unclear whether the severity of risk factors for gastric cancer such as atrophy and intestinal metaplasia are reduced after eradication in the long term. We aimed to study long-term changes in endoscopic risk factors for gastric cancer up to 20 years post-eradication. The endoscopic severity of gastritis according to the Kyoto Classification of Gastritis in 167 patients was retrospectively evaluated over an average follow-up 15.7 years. A significant improvement in mean total gastric cancer risk score (4.36 ± 1.66 to 2.69 ± 1.07, p < 0.001), atrophy (1.73 ± 0.44 to 1.61 ± 0.49, p = 0.004), and diffuse redness (1.22 ± 0.79 to 0.02 ± 0.13, p < 0.001) was observed compared to baseline in the Eradication group. However, there was no change in the never infection and current infection groups. The frequency of map-like redness increased over time until 15 years (3.6% to 18.7%, p = 0.03). The Cancer group had significantly higher risk scores at all time points. Endoscopic atrophy significantly improved in eradicated patients over long-term, suggested that eradication is one of the key elements in gastric cancer prevention. Individualized surveillance strategies based on endoscopic gastritis severity before eradication may be important for those at risk of gastric cancer.

Pathological Insights: Enhanced Vision Transformers for the Early Detection of Colorectal Cancer

Endoscopic pathological findings of the gastrointestinal tract are crucial for the early diagnosis of colorectal cancer (CRC). Previous deep learning works, aimed at improving CRC detection performance and reducing subjective analysis errors, are limited to polyp segmentation. Pathological findings were not considered and only convolutional neural networks (CNNs), which are not able to handle global image feature information, were utilized. This work introduces a novel vision transformer (ViT)-based approach for early CRC detection. The core components of the proposed approach are ViTCol, a boosted vision transformer for classifying endoscopic pathological findings, and PUTS, a vision transformer-based model for polyp segmentation. Results demonstrate the superiority of this vision transformer-based CRC detection method over existing CNN and vision transformer models. ViTCol exhibited an outstanding performance in classifying pathological findings, with an area under the receiver operating curve (AUC) value of 0.9999 ± 0.001 on the Kvasir dataset. PUTS provided outstanding results in segmenting polyp images, with mean intersection over union (mIoU) of 0.8673 and 0.9092 on the Kvasir-SEG and CVC-Clinic datasets, respectively. This work underscores the value of spatial transformers in localizing input images, which can seamlessly integrate into the main vision transformer network, enhancing the automated identification of critical image features for early CRC detection.

Comparison between the GastroPanel test and the serum pepsinogen assay interpreted with the ABC method-A prospective study

BACKGROUND AND AIMS

This study aimed to validate Helicobacter pylori serological and pepsinogen (PG) assays for detecting infection and gastric neoplasm.

METHODS

Individuals who underwent serum Chorus H. pylori and HBI PG assays were included from May to September 2023. The GastroPanel test was performed using the same blood sample. HBI assay findings were interpreted with the ABC method using the criteria of corpus atrophy (PG I ≤ 70 ng/mL & I/II ≤3) and advanced corpus atrophy (PG I ≤ 30 ng/mL & I/II ≤2).

RESULTS

A total of 144 H. pylori-infected and 184 non-infected Koreans were analyzed. The Chorus test (sensitivity 97.2%, specificity 89.1%) showed higher area under the curve (0.993 vs. 0.972, p = 0.003) than the GastroPanel test (sensitivity 95.8%, specificity 86.4%). Using the GastroSoft application, the incidence of gastric neoplasms was highest in the corpus atrophy group (50%), followed by the low acid-output (25.8%), H. pylori infection (11.6%), and antral atrophy (9.1%) groups. There were no gastric neoplasms in the normal and high acid output groups. Using the ABC method, the incidence of gastric neoplasms was highest in the corpus atrophy groups (23.8% in Groups C and D), followed by Group B (12.3%) and Group A (2.4%). Corpus atrophy interpreted with the GastroSoft showed poor agreement (k = 0.225) with corpus atrophy interpreted with the ABC method, whereas it showed excellent agreement (k = 0.854) with advanced corpus atrophy.

CONCLUSIONS

Although the Chorus test was more accurate than the GastroPanel test, both assays discriminated high-risk individuals by detecting atrophy or infection. There were no gastric neoplasms in the normal or high acid-output groups (GastroSoft application), and gastric neoplasm incidence was lowest in Group A (ABC method). Corpus atrophy determined by GastroSoft application is more consistent with advanced corpus atrophy determined by the ABC method than is corpus atrophy.

Comparison of different risk stratifications for gastric cancer and establishing a simplified risk-scoring model based on the Kyoto classification

BACKGROUND AND AIM

The study aims to assess the value of different risk stratifications in diagnosing early gastric cancer (GC) and explore risk factors based on Kyoto gastritis classification.

METHODS

This study was a single-centered cross-sectional study; all epidemiological data and endoscopic findings were obtained prospectively. To evaluate the proportion of GC in each risk stratification and to compare the diagnostic performance of different methods using the receiver operating characteristic curve, univariable and multivariable analyses were used to explore the correlation between endoscopic findings and GC.

RESULTS

A total of 240 subjects were enrolled, and the diagnostic efficacy of the Kyoto Classification Score was similar to Operative Link on Gastric Intestinal Metaplasia Assessment (OLGIM) stage, and the accuracy was higher than that of the Japanese scoring system and OLGA stage. Moderate atrophy (odds ratio [OR] = 3.52, 95% confidence interval [CI]: 1.52-8.16), severe atrophy (OR = 4.96, 95% CI: 1.75-14.04), map-like redness (OR = 9.89, 95% CI: 1.16-84.15), and xanthelasma (OR = 3.57, 95% CI: 1.15-11.15) were independent risk factors for GC. The simplified Kyoto classification (area under the receiver operating characteristic [AUROC] = 0.76, P = 0.58) based on multivariable analysis demonstrated favorable diagnostic value compared with traditional Kyoto classification score (AUROC = 0.74).

CONCLUSIONS

This study confirms the value of the Kyoto classification score and the OLGIM stage in the risk stratification of GC. Simplified Kyoto classification is also promising in risk assessment of GC but still requires validation in the population.

Gastric cancer incidence based on endoscopic Kyoto classification of gastritis

BACKGROUND

Gastric cancer (GC) incidence based on the endoscopic Kyoto classification of gastritis has not been systematically investigated using time-to-event analysis.

AIM

To examine GC incidence in an endoscopic surveillance cohort.

METHODS

This study was retrospectively conducted at the Toyoshima Endoscopy Clinic. Patients who underwent two or more esophagogastroduodenoscopies were enrolled. GC incidence was based on Kyoto classification scores, such as atrophy, intestinal metaplasia (IM), enlarged folds (EFs), nodularity, diffuse redness (DR), and total Kyoto scores. Hazard ratios (HRs) adjusted for age and sex were calculated using a Cox hazard model.

RESULTS

A total of 6718 patients were enrolled (median age 54.0 years; men 44.2%). During the follow-up period (max 5.02 years; median 2.56 years), GC developed in 34 patients. The average frequency of GCs per year was 0.19%. Kyoto atrophy scores 1 [HR with score 0 as reference: 3.66, 95% confidence interval (CI): 1.06 to 12.61], 2 (11.60, 3.82-35.27), IM score 2 (9.92, 4.37-22.54), EF score 1 (4.03, 1.63-9.96), DR scores 1 (6.22, 2.65-14.56), and 2 (10.01, 3.73-26.86) were associated with GC incidence, whereas nodularity scores were not. The total Kyoto scores of 4 (HR with total Kyoto scores 0-1 as reference: 6.23, 95%CI: 1.93 to 20.13, P = 0.002) and 5-8 (16.45, 6.29-43.03, P < 0.001) were more likely to develop GC, whereas the total Kyoto scores 2-3 were not. The HR of the total Kyoto score for developing GC per 1 rank was 1.75 (95%CI: 1.46 to 2.09, P < 0.001).

CONCLUSION

A high total Kyoto score (≥ 4) was associated with GC incidence. The endoscopy-based diagnosis of gastritis can stratify GC risk.

Cancer risk stratification system and classification of gastritis: Perspectives

Kyoto global consensus reports that the current ICD-10 classification for gastritis is obsolete. The Kyoto classification of gastritis states that severe mucosal atrophy has a high risk of gastric cancer, while mild to moderate atrophy has a low risk. The updated Kimura-Takemoto classification of atrophic gastritis considers five histological types of multifocal corpus atrophic gastritis according to stages C2 to O3. This method of morphological diagnosis of atrophic gastritis increases sensitivity by 2.4 times for severe atrophy compared to the updated Sydney system. This advantage should be considered when stratifying the high risk of gastric cancer. The updated Kimura-Takemoto classification of atrophic gastritis should be used as a reference standard (gold standard) in studies of morpho-functional relationships to identify serological markers of atrophic gastritis with evidence-based effectiveness. The use of artificial intelligence in the serological screening of atrophic gastritis makes it possible to screen a large number of the population. During serological screening of atrophic gastritis and risk stratification of gastric cancer, it is advisable to use the Kyoto classification of gastritis with updated Kimura-Takemoto classification of atrophic gastritis.

Kyoto classification of gastritis, virtual chromoendoscopy and artificial intelligence: Where are we going? What do we need?

Chronic gastritis (CG) is a widespread and frequent disease, mainly caused by Helicobacter pylori infection, which is associated with an increased risk of gastric cancer. Virtual chromoendoscopy improves the endoscopic diagnostic efficacy, which is essential to establish the most appropriate therapy and to enable cancer prevention. Artificial intelligence provides algorithms for the diagnosis of gastritis and, in particular, early gastric cancer, but it is not yet used in practice. Thus, technological innovation, through image resolution and processing, optimizes the diagnosis and management of CG and gastric cancer. The endoscopic Kyoto classification of gastritis improves the diagnosis and management of this disease, but through the analysis of the most recent literature, new algorithms can be proposed.

Decreased xCT activity in patients associated with Helicobacter pylori infection

Objective: In animals, Helicobacter pylori (Hp)-induced gastric injury is accompanied by a decrease in the activity of the cysteine/glutamate transporter (xCT), which regulates extracellular glutamate levels. However, the impact of xCT activity in patients with Hp infection remains unclear. This study aims to investigate variations of xCT activity in the gastric mucosa of patients with Hp infection and to provide a clinical basis for identifying targets related to Hp infection. Methods: Our study included a total of 67 patients with gastritis, which consisted of 44 Hp-negative and 23 Hp-positive peptic ulcer cases. The inclusion criteria used to select patients were as follows: gastric histology was determined with a gastroscope, antral biopsies were taken for urease tests, and pathology and culture were performed for analysis of Hp-colonization. The clinical characteristics of the patients were obtained, the expressions of microRNAs and xCT protein were detected using immune histochemical analysis, and the concentration of glutamate in their gastric secretion was determined. Results: The findings revealed that xCT expression was significantly lower in Hp-positive patients as compared to Hp-negative individuals, which was accompanied by a decrease in glutamate concentration in gastric juice. We also discovered a high expression of microRNAs that have been shown to negatively regulate xCT expression, in Hp-positive patients. Conclusion: Reduced xCT activity in patients may play an important role in gastric ulcers caused by Hp infection. Our findings suggest that the microRNA/xCT pathway could be a potential treatment target for Hp-infection-related ulcers.

Kyoto classification of gastritis: Advances and future perspectives in endoscopic diagnosis of gastritis

This editorial provides an update of the recent evidence on the endoscopy-based Kyoto classification of gastritis, clarifying the shortcomings of the Kyoto classification, and providing prospects for future research, with particular focus on the histological subtypes of gastric cancer (GC) and Helicobacter pylori (H. pylori) infection status. The total Kyoto score is designed to express GC risk on a score ranging from 0 to 8, based on the following five endoscopic findings: Atrophy, intestinal metaplasia (IM), enlarged folds (EF), nodularity, and diffuse redness (DR). The total Kyoto score reflects H. pylori status as follows: 0, ≥ 2, and ≥ 4 indicate a normal stomach, H. pylori-infected gastritis, and gastritis at risk for GC, respectively. Regular arrangement of collecting venules (RAC) predicts non-infection; EF, nodularity, and DR predict current infection; map-like redness (MLR) predicts past infection; and atrophy and IM predict current or past infection. Atrophy, IM, and EF all increase the incidence of H. pylori-infected GC. MLR is a specific risk factor for H. pylori-eradicated GC, while RAC results in less GC. Diffuse-type GC can be induced by active inflammation, which presents as EF, nodularity, and atrophy on endoscopy, as well as neutrophil and mononuclear cell infiltration on histology. In contrast, intestinal-type GC develops via atrophy and IM, and is consistent between endoscopy and histology. However, this GC risk-scoring design needs to be improved.

Current and Future Perspectives in the Diagnosis and Management of Helicobacter pylori Infection

Helicobacter pylori (Hp) is a prevalent organism infecting almost half the global population. It is a significant concern, given its associated risk of gastric cancer, which is the third leading cause of cancer death globally. Infection can be asymptomatic or present with dyspeptic symptoms. It may also present with alarm symptoms in the case of progression to cancer. Diagnosis can be achieved non-invasively (breath tests, stool studies, or serology) or invasively (rapid urease test, biopsy, or culture). Treatment involves acid suppression and regimens containing several antibiotics and is guided by resistance rates. Eradication is essential, as it lowers the risk of complications and progression to cancer. Follow-up after eradication is similarly important, as the risk of cancer progression remains. There have been many recent advances in both diagnosis and treatment of Hp. In particular, biosensors may be effective diagnostic tools, and nanotechnology, vaccines, and potassium-competitive acid blockers may prove effective in enhancing eradication rates.