How commonly is upper gastrointestinal cancer missed at endoscopy? A meta-analysis

Endoscopic detection and diagnosis of gastric cancer using image-enhanced endoscopy: A systematic review and meta-analysis

Objectives

We aimed to conduct a systematic review and meta-analysis to assess the value of image-enhanced endoscopy including blue laser imaging (BLI), linked color imaging, narrow-band imaging (NBI), and texture and color enhancement imaging to detect and diagnose gastric cancer (GC) compared to that of white-light imaging (WLI).

Methods

Studies meeting the inclusion criteria were identified through PubMed, Cochrane Library, and Japan Medical Abstracts Society databases searches. The pooled risk ratio for dichotomous variables was calculated using the random-effects model to assess the GC detection between WLI and image-enhanced endoscopy. A random-effects model was used to calculate the overall diagnostic performance of WLI and magnifying image-enhanced endoscopy for GC.

Results

Sixteen studies met the inclusion criteria. The detection rate of GC was significantly improved in linked color imaging compared with that in WLI (risk ratio, 2.20; 95% confidence interval [CI], 1.39-3.25; p < 0.01) with mild heterogeneity. Magnifying endoscopy with NBI (ME-NBI) obtained a pooled sensitivity, specificity, and area under the summary receiver operating curve of 0.84 (95 % CI, 0.80-0.88), 0.96 (95 % CI, 0.94-0.97), and 0.92, respectively. Similarly, ME-BLI showed a pooled sensitivity, specificity, and area under the curve of 0.81 (95 % CI, 0.77-0.85), 0.85 (95 % CI, 0.82-0.88), and 0.95, respectively. The diagnostic efficacy of ME-NBI/BLI for GC was evidently high compared to that of WLI, However, significant heterogeneity among the NBI studies still existed.

Conclusions

Our meta-analysis showed a high detection rate for linked color imaging and a high diagnostic performance of ME-NBI/BLI for GC compared to that with WLI.

GastroHUN an Endoscopy Dataset of Complete Systematic Screening Protocol for the Stomach

Endoscopy is vital for detecting and diagnosing gastrointestinal diseases. Systematic examination protocols are key to enhancing detection, particularly for the early identification of premalignant conditions. Publicly available endoscopy image databases are crucial for machine learning research, yet challenges persist, particularly in identifying upper gastrointestinal anatomical landmarks to ensure effective and precise endoscopic procedures. However, many existing datasets have inconsistent labeling and limited accessibility, leading to biased models and reduced generalizability. This paper introduces GastroHUN, an open dataset documenting stomach screening procedures based on a systematic protocol. GastroHUN includes 8,834 images from 387 patients and 4,729 labeled video sequences, all annotated by four experts. The dataset covers 22 anatomical landmarks in the stomach and includes an additional category for unqualified images, making it a valuable resource for AI model development. By providing a robust public dataset and baseline deep learning models for image and sequence classification, GastroHUN serves as a benchmark for future research and aids in the development of more effective algorithms.

The variation in post-endoscopy upper gastrointestinal cancer rates among endoscopy providers in England and associated factors: a population-based study

BACKGROUND

Post-endoscopy upper gastrointestinal cancer (PEUGIC) is an important key performance indicator for endoscopy quality. We examined variation in PEUGIC rates among endoscopy providers in England and explored associated factors.

METHODS

The was a population-based, retrospective, case-control study, examining data from National Cancer Registration and Analysis Service and Hospital Episode Statistics databases for esophageal and gastric cancers diagnosed between 2009 and 2018 in England. PEUGIC were cancers diagnosed 6 to 36 months after an endoscopy that did not diagnose cancer. Associated factors were identified using multivariable logistic regression analyses.

RESULTS

The national PEUGIC rate was 8.5%, varying from 5% to 13% among endoscopy providers. Factors associated with PEUGIC included: female sex (odds ratio [OR] 1.29 [95%CI 1.23-1.36]); younger age (age >80 years, OR 0.52 [0.48-0.56], compared with ≤60 years); increasing comorbidity (Charlson score >4, OR 5.06 [4.45-5.76]); history of esophageal ulcer (OR 3.30 [3.11-3.50]), Barrett's esophagus (OR 3.21 [3.02-3.42]), esophageal stricture (OR 1.28 [1.20-1.37]), or gastric ulcer (OR 1.55 [1.44-1.66]); squamous cell carcinoma (OR 1.50 [1.39-1.61]); and UK national endoscopy accreditation status - providers requiring improvement (OR 1.10 [1.01-1.20]), providers never assessed (OR 1.24 [1.04-1.47]).

CONCLUSION

PEUGIC rates varied threefold among endoscopy providers, suggesting unwarranted differences in endoscopy quality. PEUGIC was associated with endoscopy findings known to be associated with upper gastrointestinal cancer and a lack of national endoscopy provider accreditation. PEUGIC variations suggest an opportunity to raise performance standards to detect upper gastrointestinal cancers earlier and improve outcomes.

The endoscopic model for gastric carcinogenesis and Helicobacter pylori infection: A potential visual mind-map during gastroscopy examination

Helicobacter pylori (Hp) is the main trigger of chronic gastric atrophy and the main leading cause of gastric cancer. Hp infects the normal gastric mucosa and can lead to chronic inflammation, glandular atrophy, intestinal metaplasia, dysplasia and finally adenocarcinoma. Chronic inflammation and gastric atrophy associated with Hp infection appear initially in the distal part of the stomach (the antrum) before progressing to the proximal part (the corpus-fundus). In recent years, endoscopic developments have allowed for the characterization of various gastric conditions including the normal mucosa (pyloric/fundic gland pattern and regular arrangement of collecting venules), Hp-related gastritis (Kyoto classification), glandular atrophy (Kimura-Takemoto classification), intestinal metaplasia (Endoscopic Grading of Gastric Intestinal Metaplasia), and dysplasia/adenocarcinoma (Vessel plus Surface classification). Despite being independent classifications, all these scales can be integrated into a single model: the endoscopic model for gastric carcinogenesis. This model would assist endoscopists in comprehending the process of gastric carcinogenesis and conducting a systematic examination during gastroscopy. Having this model in mind would enable endoscopists to promptly recognize the implications of Hp infection and the potential patient's risk of developing gastric cancer.

Validation of Artificial Intelligence Computer-Aided Detection on Gastric Neoplasm in Upper Gastrointestinal Endoscopy

BACKGROUND/OBJECTIVES

Gastric cancer ranks fifth for incidence and fourth in the leading causes of mortality worldwide. In this study, we aimed to validate previously developed artificial intelligence (AI) computer-aided detection (CADe) algorithm, called ALPHAON® in detecting gastric neoplasm.

METHODS

We used the retrospective data of 500 still images, including 5 benign gastric ulcers, 95 with gastric cancer, and 400 normal images. Thereby we validated the CADe algorithm measuring accuracy, sensitivity, and specificity with the result of receiver operating characteristic curves (ROC) and area under curve (AUC) in addition to comparing the diagnostic performance status of four expert endoscopists, four trainees, and four beginners from two university-affiliated hospitals with CADe algorithm. After a washing-out period of over 2 weeks, endoscopists performed gastric detection on the same dataset of the 500 endoscopic images again marked by ALPHAON®.

RESULTS

The CADe algorithm presented high validity in detecting gastric neoplasm with accuracy (0.88, 95% CI: 0.85 to 0.91), sensitivity (0.93, 95% CI: 0.88 to 0.98), specificity (0.87, 95% CI: 0.84 to 0.90), and AUC (0.962). After a washing-out period of over 2 weeks, overall validity improved in the trainee and beginner groups with the assistance of ALPHAON®. Significant improvement was present, especially in the beginner group (accuracy 0.94 (0.93 to 0.96) p < 0.001, sensitivity 0.87 (0.82 to 0.92) p < 0.001, specificity 0.96 (0.95 to 0.97) p < 0.001).

CONCLUSIONS

The high validation performance state of the CADe algorithm system was verified. Also, ALPHAON® has demonstrated its potential to serve as an endoscopic educator for beginners improving and making progress in sensitivity and specificity.

Clinical application of capsule sponge testing in symptomatic reflux disease: a national prospective cohort study

BACKGROUND

Capsule sponge testing, using an oesophageal cell collection device with biomarkers, was implemented nationally across Scotland in 2020 for symptomatic reflux patients referred to secondary care for non-urgent endoscopy. The aim was to use capsule sponge testing as a triage tool to reduce pressures on the endoscopy service during COVID-19, focus endoscopy resources on those most likely to have pathology and streamline the patient pathway. This prospective cohort study presents the first real-world results and evaluates the clinical application of capsule sponge testing in symptomatic reflux disease based on endoscopic biopsy results.

METHODS

Over a 32-month period, all patients undergoing capsule sponge testing for investigation of reflux symptoms across 11 Scottish health boards were identified from prospectively maintained databases. Individual patient records were interrogated to collect baseline demographics, capsule sponge test result (TFF3/atypia/p53) and ongoing clinical management. Further analysis was performed on patients who subsequently underwent upper gastrointestinal (UGI) endoscopy.

RESULTS

1385 tests were performed for reflux symptoms in 1305 patients. The median follow-up time was 20 months (IQR 12-27). 1103/1385 tests (79.6%) were negative for biomarkers. 913/1305 (70.0%) patients were discharged with no additional investigation required. 355/1305 patients (27.2%) underwent UGI endoscopy due to a positive or insufficient result or ongoing symptoms. With insufficient tests excluded, 52/314 patients (16.6%) had intestinal metaplasia (IM) on endoscopic biopsies, which strongly correlated with positive biomarkers (88.5% vs. 11.5%; p < 0.001), including 1 case of dysplasia. 10/1103 sponge negative patients (0.9%) had biopsies demonstrating IM (n = 6) or malignancy (n = 4): 1 patient was diagnosed with oesophageal adenocarcinoma 27 months later and 3 patients had gastric malignancy, which relies on symptom assessment to direct to endoscopy since the capsule sponge is an oesophageal test.

CONCLUSIONS

Capsule sponge testing is effectively at identifying the 30% of symptomatic reflux patients requiring further investigation with UGI endoscopy and aiding the diagnosis of Barrett's oesophagus in clinical practice. Judicious follow-up of the discharged group will be critical to validate the ongoing use of capsule sponge testing long-term.

Discharging the duty of candor following delayed post-endoscopy cancer diagnosis

Endoscopic examination is not risk free. Not only are there well-known complications associated with the procedure, but malignant and pre-malignant lesions can be missed due to human factors or failures in organizational process. Duty of candor (DoC) is a legal requirement if significant harm occurs in delivery of healthcare. Post-colonoscopy colorectal cancer (PCCRC) and post-endoscopy upper gastrointestinal cancer (PEUGIC) audits have identified missed diagnoses that are associated with harm and require consideration of DoC. This article explores the new and unique challenges associated with DoC in endoscopy audits. There are unresolved questions around the place of DoC in retrospective audits, agreement of harm thresholds, and constitution of review teams. Involved departments must be committed to transparency and trained in governance processes. Fear of institutional and personal reputational damage, as well as future litigation, may influence decisions. Patient expectations need to be clarified, as do supportive structures for individual endoscopists who will be involved in DoC processes when significant lesions have been missed. Further consensus around DoC is required so that clear guidance can be given to endoscopy units.

Confocal laser endomicroscopy as a new diagnostic tool for poorly differentiated gastric adenocarcinoma

Gastric cancer (GC) is a multifactorial disease, where both environmental and genetic features can have an impact on its occurrence and development. GC represents one of the leading causes of cancer-related deaths worldwide. GC is most frequent in males and is believed to arise from a series of premalignant lesions. The detection of GC at an early stage is crucial because early GC, which is an invasive stomach cancer confined to the mucosal or submucosal lining, may be curable with a reported 5-year survival rate of more than 90%. Advanced GC usually has a poor prognosis despite current treatment standards. The diagnostic efficacy of conventional endoscopy (with light endoscopy) is currently limited. Confocal laser endomicroscopy is a novel imaging technique that allows real-time in vivo histological examination of mucosal surfaces during endoscopy. Confocal laser endomicroscopy may be of great importance in the surveillance of precancerous gastric lesions and in the diagnosis of GC. In this editorial we commented on the article about this topic published by Lou et al in the recent issue of the World Journal of Clinical Cases.

Gastric Cancer and Intestinal Metaplasia: Differential Metabolic Landscapes and New Pathways to Diagnosis

Gastric cancer (GC) is the fifth most common cause of cancer-related death worldwide. Early detection is crucial for improving survival rates and treatment outcomes. However, accurate GC-specific biomarkers remain unknown. This study aimed to identify the metabolic differences between intestinal metaplasia (IM) and GC to determine the pathways involved in GC. A metabolic analysis of IM and tissue samples from 37 patients with GC was conducted using ultra-performance liquid chromatography with tandem mass spectrometry. Overall, 665 and 278 significant features were identified in the aqueous and 278 organic phases, respectively, using false discovery rate analysis, which controls the expected proportion of false positives among the significant results. sPLS-DA revealed a clear separation between IM and GC samples. Steroid hormone biosynthesis, tryptophan metabolism, purine metabolism, and arginine and proline metabolism were the most significantly altered pathways. The intensity of 11 metabolites, including N1, N2-diacetylspermine, creatine riboside, and N-formylkynurenine, showed significant elevation in more advanced GC. Based on pathway enrichment analysis and cancer stage-specific alterations, we identified six potential candidates as diagnostic biomarkers: aldosterone, N-formylkynurenine, guanosine triphosphate, arginine, S-adenosylmethioninamine, and creatine riboside. These metabolic differences between IM and GC provide valuable insights into gastric carcinogenesis. Further validation is needed to develop noninvasive diagnostic tools and targeted therapies to improve the outcomes of patients with GC.

Protocol for a national, multicentre study of post-endoscopy colorectal and upper gastrointestinal cancers: The POET study

AIM

The primary aim of the study is to define the post-colonoscopy colorectal cancer (PCCRC) three-year rate and the post-endoscopy upper gastrointestinal cancer (PEUGIC) three-year rate across public hospitals in Aotearoa New Zealand.

METHOD

This retrospective cohort study will be conducted via the trainee-led STRATA Collaborative network. All public hospitals in Aotearoa New Zealand will be eligible to participate. Data will be collected on all adult patients who are diagnosed with colorectal adenocarcinoma within 6 to 48 months of a colonoscopy and all adult patients diagnosed with gastroesophageal cancer within 6 to 48 months of an upper gastrointestinal endoscopy. The study period will be from 2010 to 2022. The primary outcome is the PCCRC 3-year rate and the PEUGIC 3-year rate. Secondary aims are to define and characterize survival after PCCRC or PEUGIC, the cause of PCCRC as based on the World Endoscopy Organization System of Analysis definitions, trends over time, and centre level variation.

CONCLUSION

This protocol describes the methodology for a nationwide retrospective cohort study on PCCRC and PEUGIC in Aotearoa New Zealand. These data will lay the foundation for future studies and quality improvement initiatives.

Efficacy of a real-time intelligent quality-control system for the detection of early upper gastrointestinal neoplasms: a multicentre, single-blinded, randomised controlled trial

Background

Oesophagogastroduodenoscopy (OGD) quality and identification of the early upper gastrointestinal (UGI) neoplasm play an important role in detecting the UGI neoplasm. However, the optimal method for quality control in daily OGD procedures is currently lacking. We aimed to evaluate the efficacy of a real-time intelligent quality-control system (IQCS), which combines OGD quality control with lesion detection of early UGI neoplasms.

Methods

We performed a multicentre, single-blinded, randomised controlled trial at 6 hospitals in China. Patients aged 40-80 years old who underwent painless OGD were screened for enrolment in this study. Patients with a history of advanced UGI cancer, stenosis, or obstruction in UGI tract were excluded. Eligible subjects were randomly assigned (1:1) to either the routine or IQCS group to undergo standard OGD examination and OGD examination aided by IQCS, respectively. Patients were masked to the randomisation status. The primary outcome was the detection of early UGI neoplasms. All analyses were done on a per-protocol basis. This trial is registered with ClinicalTrials.gov, NCT04720924.

Findings

Between January 16, 2021 and December 23, 2022, 1840 patients were randomised (IQCS group: 919, routine group: 921). The full analysis set consisted of 914 in the IQCS group and 915 in the routine group. The early UGI neoplasms detection rate in the IQCS group (6.1%, 56/914) was significantly higher than in the routine group (2.3%, 21/915; P = 0.0001). The IQCS group had fewer blind spots (2.3 vs. 6.2, P < 0.0001). The IQCS group had higher stomach cleanliness on cardia or fundus (99.5% vs. 87.9%, P < 0.0001), body (98.9% vs. 88.0%, P < 0.0001), angulus (99.8% vs. 88.4%, P < 0.0001) and antrum or pylorus (100.0% vs. 87.4%, P < 0.0001). The inspection time (576.2 vs. 574.5s, P = 0.91) and biopsy rate (57.2% vs. 56.6%, P = 0.83) were not different between the groups. The early UGI neoplasms detection rate in the IQCS group increased in both non-academic centres (RR = 3.319, 95% CI 1.277-9.176; P = 0.0094) and academic centres (RR = 2.416, 95% CI 1.301-4.568; P = 0.0034). The same improvements were observed for both less-experienced endoscopists (RR = 2.650, 95% CI 1.330-5.410; P = 0.0034) and experienced endoscopists (RR = 2.710, 95% CI 1.226-6.205; P = 0.010). No adverse events or serious adverse events were reported in the two groups.

Interpretation

The IQCS improved the OGD quality and increased early UGI neoplasm detection in different hospital types and endoscopist experiences. IQCS could play an important role in primary basic hospitals and non-expert endoscopists to improve the diagnostic accuracy of early UGI neoplasms. The effectiveness of IQCS in real-world clinical settings needs a larger population validation.

Funding

Key R&D Program of Shandong Province, China (Major Scientific and Technological Innovation Project), National Natural Science Foundation of China, the Taishan Scholars Program of Shandong Province, the National Key Research and Development Program of China, and the Shandong Provincial Natural Science Foundation.

Global epidemiology of early-onset upper gastrointestinal cancer: trend from the Global Burden of Disease Study 2019

BACKGROUND AND AIM

In recent years, there has been a growing incidence of gastrointestinal cancer in young individuals. Despite its significant morbidity and mortality, research on upper gastrointestinal (UGI) cancer in young populations has been relatively limited. Therefore, studies on the epidemiological changes of this cancer are needed.

METHODS

Using data from the Global Burden of Disease Study 2019, we examined the incidence, death, and disability-adjusted life years (DALYs) from UGI cancers in the young, namely, early-onset esophageal cancer (EOEC) and early-onset gastric cancer (EOGC). These results were stratified by sex, geographical region, country, and sociodemographic index.

RESULTS

There was a total of 185 140 cases, 120 289 deaths, and 5.70 million DALYs attributable to early-onset UGI cancers globally. From 2010 to 2019, the global incidence, death, and DALYs rates of early-onset UGI cancers decreased. In contrast, the incidence rates increased in both EOEC (+1.15%) and EOGC (+0.21%) in the Eastern Mediterranean region.

CONCLUSIONS

Over the past decade, the burden of UGI cancer in the young has decreased. However, it has increased in the Eastern Mediterranean region. Further research to elucidate the attributable risk factors in this population is warranted.

Linked Color Imaging with Artificial Intelligence Improves the Detection of Early Gastric Cancer

INTRODUCTION

Esophagogastroduodenoscopy is the most important tool to detect gastric cancer (GC). In this study, we developed a computer-aided detection (CADe) system to detect GC with white light imaging (WLI) and linked color imaging (LCI) modes and aimed to compare the performance of CADe with that of endoscopists.

METHODS

The system was developed based on the deep learning framework from 9,021 images in 385 patients between 2017 and 2020. A total of 116 LCI and WLI videos from 110 patients between 2017 and 2023 were used to evaluate per-case sensitivity and per-frame specificity.

RESULTS

The per-case sensitivity and per-frame specificity of CADe with a confidence level of 0.5 in detecting GC were 78.6% and 93.4% for WLI and 94.0% and 93.3% for LCI, respectively (p < 0.001). The per-case sensitivities of nonexpert endoscopists for WLI and LCI were 45.8% and 80.4%, whereas those of expert endoscopists were 66.7% and 90.6%, respectively. Regarding detectability between CADe and endoscopists, the per-case sensitivities for WLI and LCI were 78.6% and 94.0% in CADe, respectively, which were significantly higher than those for LCI in experts (90.6%, p = 0.004) and those for WLI and LCI in nonexperts (45.8% and 80.4%, respectively, p < 0.001); however, no significant difference for WLI was observed between CADe and experts (p = 0.134).

CONCLUSIONS

Our CADe system showed significantly better sensitivity in detecting GC when used in LCI compared with WLI mode. Moreover, the sensitivity of CADe using LCI is significantly higher than those of expert endoscopists using LCI to detect GC.

Disease burden and related risk factors of esophageal cancer in China and globally from 1990 to 2021, with forecast to 2035: An analysis and comparison

INTRODUCTION

In this study we estimate the burden of esophageal cancer (EC) in China and globally from 1990 to 2021, with a forecast to 2035, using Global Burden of Disease (GBD) data. We also analyze the related risk factors to investigate burden trends.

METHODS

Mortality, disability-adjusted life years (DALYs), crude rates, and age-standardized rates of EC were analyzed in China and globally from 1990 to 2035, utilizing GBD open data as a secondary dataset analysis of GBD data. Temporal change trends of EC risk factors were analyzed from 1990 to 2021. Joinpoint regression determined average annual percentage change (AAPC) of age-standardized rates. Descriptive analysis compared mortality and DALYs by age groups. Bayesian age-period-cohort (BAPC) predicted age-standardized mortality and DALYs rates for the next 14 years.

RESULTS

The ASMR and ASDR fluctuations in EC were significant in China, showing an overall downward trend. Globally, although there was also a downward trend, the fluctuations were relatively mild. The number of deaths and DALYs related to EC in China and globally showed a significant upward trend. Age-specific burden trends in China for EC indicated that the age group with the peak number of EC deaths shifted to the 70-74 years age group in 2021, while DALYs peaked in the 65-69 years age group. The crude mortality rate (CMR) peaked consistently in 1990 and 2021, both within the 90-94 years age range, while the crude DALY rate (CDR) shifted to the 85-89 years age group. Overall, the burden of EC deaths and DALYs in the population aged <40 years was relatively low, increasing rapidly after the age of 40 years, reaching a peak and gradually declining, and reaching a lower level after the age of 85 years. The predictive results of the BAPC model indicated that over the next 14 years, both ASMR and ASDR for EC in China and globally would show a slight overall increase. The GBD 2021 study identified smoking, high alcohol use, chewing tobacco, and diet low in vegetables as the main risk factors affecting EC mortality rate and DALYs. Among these, smoking and alcohol use were the most significant risk factors, with a higher impact on EC in China compared to the global level. From 1990 to 2021, the overall changes in ASMR and ASDR indicate a decreasing trend in the impact of these four risk factors on EC mortality rate and DALYs.

CONCLUSIONS

The burden of EC is expected to steadily increase in China and globally until 2035, posing a significant challenge. Targeted prevention and control policies, such as calling on people to quit smoking and reduce alcohol use, may help curb this upward trend.

Association between anesthesia assistance and precancerous lesions and early cancer detection during diagnostic esophagogastroduodenoscopy: a propensity score-matched retrospective study

Background

Esophagogastroduodenoscopy (EGD) is a fundamental procedure for early detection of upper gastrointestinal (UGI) cancer. However, limited research has been conducted on the impact of sedation during EGD on the identification of precancerous lesions and early cancer (EC). This retrospective study aims to evaluate whether sedation during EGD can improve the detection rates of precancerous lesions and EC.

Methods

In this propensity score-matched retrospective study, we examined medical records from outpatients who underwent diagnostic EGD at a large tertiary center between January 2023 and December 2023. Data on endoscopic findings and histology biopsies were obtained from an endoscopy quality-control system. The primary objective was to compare the rates of detecting precancerous lesions and EC in patients who received sedation during EGD vs. those who did not receive sedation. Additionally, we aimed to identify factors influencing these detection rates using binary logistic regression analysis.

Results

Following propensity score matching, a total of 17,862 patients who underwent diagnostic EGD with or without propofol sedation were identified. The group that received sedation exhibited a higher detection rate of precancerous lesions and EC in comparison to the non-sedated group (1.04 vs. 0.75%; p = 0.039). Additionally, within the sedated group, there was an increased likelihood of identifying precancerous lesions and EC specifically at the gastric antrum (0.60 vs. 0.32%, p = 0.006). Binary logistic regression analysis demonstrated that independent risk factors influencing the detection rates included age, gender, observation time, and number of biopsies conducted during the procedure.

Conclusion

Anesthesia assistance during EGD screening proved advantageous in detecting EC as well as precancerous lesions. It is crucial for endoscopists to consider these factors when performing EGD screening procedures.

Prognostic impact of deprivation on esophagogastroduodenoscopy outcome

Background and study aims Socioeconomic deprivation has long been associated with many gastrointestinal diseases, yet its influence on esophagogastroduodenoscopy (EGD) diagnosis has not been evaluated. The aim of this study was to investigate the influence of deprivation on outcomes of EGD irrespective of referral reason. Patients and methods Two thousand consecutive patients presenting to four Health Boards in Wales beginning in June 2019 were studied retrospectively with deprivation scores calculated using the Wales Indices of Multiple Deprivation (WIMD). Patients were subclassified into quintiles for analysis (Q1 most, Q5 least deprived). Results Inhabitants of the most deprived areas were more likely to be diagnosed with peptic ulcer (Q1 7.9%, Q5 4.7%; odds ratio [OR] 0.498, P =0.018), severe esophagitis (LA4, Q1 2.7% v Q5 0%, OR 0.089, P 0.002), Helicobacter pylori infection (Q1 5.4%, Q5 1.7%; OR 0.284, P =0.002), but less likely to be diagnosed with Barrett's esophagus (Q1 6.3% v Q5 12.3%, OR 2.146, P =0.004) than those from the least deprived areas. New cancer diagnoses numbered 53 and were proportionately higher after presentation for urgent suspected cancer (USC, n=35, 4.6%) than for routine referrals (n=3, 0.6%, P < 0.001). Deprivation was associated with more advanced stage cancer (stage III Q1 16.7% v Q5 5.6%, OR 0.997, P =0.006: stage IV Q1 16.7% v Q2 38.9% v Q5 22.2%, OR 0.998, P =0.049). Conclusions Deprivation was associated with two-fold more peptic ulcer disease, three-fold more H. pylori infection, and 12-fold more severe esophagitis, and more advanced cancer stage.

Use of N-acetylcysteine plus simethicone to improve mucosal visibility in upper digestive endoscopy via systematic alphanumeric-coded endoscopy: a randomized, double-blind controlled trial

Background

The use of antifoaming and mucolytic agents prior to upper gastrointestinal (GI) endoscopy and a thorough systematic review are essential to optimize lesion detection. This study evaluated the effect of simethicone and N-acetylcysteine on the adequate mucosal visibility (AMV) of the upper GI tract by an innovative systematic method.

Methods

This randomized, double-blind controlled trial included consecutive patients who underwent diagnostic upper GI endoscopy for screening for early neoplasms between August 2019 and December 2019. The upper GI tract was systematically assessed by systematic alphanumeric-coded endoscopy. Patients were divided into 4 groups: 1) water; 2) only simethicone; 3) N-acetylcysteine + simethicone; and 4) only N-acetylcysteine. The following parameters were assessed in each group: age, sex, body mass index, level of adequate mucosal visibility, and side-effects.

Results

A total of 4564 images from upper GI areas were obtained for evaluation. The mean AMV in the 4 groups was 93.98±7.36%. The N-acetylcysteine + simethicone group had a higher cleaning percentage compared with the other groups (P=0.001). There was no significant difference among the remaining groups, but several areas had better cleaning when a mucolytic or antifoam alone was used. No side-effects were found in any group.

Conclusion

The combination of N-acetylcysteine plus simethicone optimizes the visibility of the mucosa of the upper GI tract, which could potentially increase diagnostic yield.

The durability of previous examinations for cancer: Danish nationwide cohort study

OBJECTIVE

Patients previously examined for cancer with a negative result may present in general practice with ongoing or new symptoms or signs suggestive of cancer. This paper explores the potential existence of a relatively safe period for cancer occurrence after receiving negative examination results for specific types of cancer, including lung (CT thorax), upper gastrointestinal (gastroscopy), colorectal (colonoscopy), bladder (cystoscopy), and breast (clinical mammography).

DESIGN

Register-based time-to-event analyses.

SETTING

Denmark.

SUBJECTS

All 3.3 million citizens aged 30-85 years who on January first, 2017, had not previously been diagnosed with the specific type of cancer were categorized based on the time since their most recent examination.

MAIN OUTCOME MEASURES

Using 1-year follow-up, we calculated the age- and sex-adjusted hazard ratios of being diagnosed with the related cancer, with non-examined individuals as reference. Negative examination results were defined as the absence of a cancer diagnosis within 6 months following the examination.

RESULTS

Previous negative examination results were common, also among those diagnosed with cancer during follow-up. For 10 years after a negative colonoscopy the risk of diagnosing a colorectal cancer was nearly halved. However, already 1 year after a clinical mammography and 2 years after a CT thorax the risk of diagnosing the related cancers was significantly higher among those with a previous negative result compared to non-examined individuals.

CONCLUSION

This study did not identify a post-examination period in which the cancer risk, compared to non-examined individuals, was sufficiently low to confidently rule out any of the investigated cancers.

Enhanced multi-class pathology lesion detection in gastric neoplasms using deep learning-based approach and validation

This study developed a new convolutional neural network model to detect and classify gastric lesions as malignant, premalignant, and benign. We used 10,181 white-light endoscopy images from 2606 patients in an 8:1:1 ratio. Lesions were categorized as early gastric cancer (EGC), advanced gastric cancer (AGC), gastric dysplasia, benign gastric ulcer (BGU), benign polyp, and benign erosion. We assessed the lesion detection and classification model using six-class, cancer versus non-cancer, and neoplasm versus non-neoplasm categories, as well as T-stage estimation in cancer lesions (T1, T2-T4). The lesion detection rate was 95.22% (219/230 patients) on a per-patient basis: 100% for EGC, 97.22% for AGC, 96.49% for dysplasia, 75.00% for BGU, 97.22% for benign polyps, and 80.49% for benign erosion. The six-class category exhibited an accuracy of 73.43%, sensitivity of 80.90%, specificity of 83.32%, positive predictive value (PPV) of 73.68%, and negative predictive value (NPV) of 88.53%. The sensitivity and NPV were 78.62% and 88.57% for the cancer versus non-cancer category, and 83.26% and 89.80% for the neoplasm versus non-neoplasm category, respectively. The T stage estimation model achieved an accuracy of 85.17%, sensitivity of 88.68%, specificity of 79.81%, PPV of 87.04%, and NPV of 82.18%. The novel CNN-based model remarkably detected and classified malignant, premalignant, and benign gastric lesions and accurately estimated gastric cancer T-stages.

Efficacy of Cricoid Pressure Application during Esophagogastroduodenoscopy in Patients with Poor Gastric Wall Extension

INTRODUCTION

Esophagogastroduodenoscopy (EGD) requires adequate air infusion. However, cases of poor gastrointestinal wall extension due to frequent eructation have been reported. Sufficient gastrointestinal wall extension can be achieved by applying cricoid pressure during EGD. Herein, we evaluated the frequency of cases with poor gastrointestinal wall extension and the efficacy and safety of applying cricoid pressure during EGD.

METHODS

This interventional study included patients who underwent EGD between January 2020 and December 2020 at the JA Akita Koseiren Yuri Kumiai General Hospital. Cases wherein folds of the greater curvature of the upper gastric body were not sufficiently extended during EGD were considered to have poor gastrointestinal wall extension. In such cases, air infusion was performed while applying cricoid pressure. This procedure was considered effective when gastric wall extension was achieved.

RESULTS

A total of 2,000 patients were enrolled and underwent upper gastrointestinal endoscopy; however, five were excluded because of upper gastrointestinal tract stenosis. Observation of gastric wall extension of the greater curvature in the upper gastric body with normal air insufflation was difficult in 113 (5.7%) cases. Applying cricoid pressure was effective in 93 (82.3%) patients with poor gastric wall extension. Sufficient gastric wall extension was achieved within an average of 12.8 s in cases where cricoid pressure application was effective. No adverse events were associated with cricoid pressure application.

CONCLUSIONS

Cricoid pressure application for patients with poor gastric wall extension during EGD is useful for ensuring a sufficient field of view during observation of the gastric body.

New Horizons of Artificial Intelligence in Medicine and Surgery

Background: Ideas about Artificial intelligence appeared about half a century ago, but only now is it becoming an essential element of everyday life. The data provided are becoming a bigger pool and we need artificial intelligence that will help us with its superhuman powers. Its interaction with medicine is improving more and more, with medicine being a domain that continues to be perfected. Materials and Methods: The most important databases were used to perform this detailed search that addresses artificial intelligence in the medical and surgical fields. Discussion: Machine learning, deep learning, neural networks and computer vision are some of the mechanisms that are becoming a trend in healthcare worldwide. Developed countries such as Japan, France and Germany have already implemented artificial intelligence in their medical systems. The help it gives is in medical diagnosis, patient monitoring, personalized therapy and workflow optimization. Artificial intelligence will help surgeons to perfect their skills, to standardize techniques and to choose the best surgical techniques. Conclusions: The goal is to predict complications, reduce diagnostic times, diagnose complex pathologies, guide surgeons intraoperatively and reduce medical errors. We are at the beginning of this, and the potential is enormous, but we must not forget the impediments that may appear and slow down its implementation.

Endoscopic diagnosis and prevalence of early gastric cancer in India: A prospective study

Objectives

Although countries like Japan and South Korea have implemented nationwide endoscopic screening programs, there is limited evidence on the effectiveness of endoscopy for diagnosing early gastric cancer (EGC) in developing countries such as India. In the present study, we aimed to determine the feasibility of endoscopic detection of EGC from India.

Methods

The data was prospectively collected for all patients ≥40 years who underwent a diagnostic upper endoscopy from April to September 2021. A single endoscopist who performed the endoscopic procedures completed 1-month training in advanced endoscopy in Japan. Following the training, the endoscopist continued to engage in internet-based discussions regarding his cases encountered. Prior to this training, the endoscopist had not detected any EGC cases during his 12-year gastroenterology practice.

Results

A total of 1033 patients were included in the study, with males accounting for 65.4% and a mean age 52 years. The average procedural time was 7.13 ± 4.8 min. A total of 25 patients (2.4%) were found to have GC, including 6 patients (0.6%) with EGC. Two patients had synchronous EGC lesions. All EGC patients were males, with an average age of 66 years. All EGCs were detected in the distal stomach in the presence of Helicobacter pylori infection and severe atrophic gastritis.

Conclusion

Our findings showed that the endoscopic detection of EGC is feasible in India. Optimal training on endoscopic diagnosis of EGC can improve the detection of such lesion. Further studies are warranted to assess the optimization and implementation of an endoscopic screening program for EGC in India.

Impact of endoscopic surveillance on the early diagnosis and endoscopic resection likelihood of gastric cancer

BACKGROUND

Endoscopy could help detect early gastric cancer (EGC) and improve the prognosis of patients. The aim of this study was to analyze the impact of endoscopy and endoscopic surveillance on the early detection of gastric cancer (GC), GC staging, and treatment selection.

METHODS

Patients with GC diagnosed at our center from 2010 to 2022 were retrospectively analyzed and allocated to the short-interval group (had received endoscopy within 3 years before diagnosis), the long-interval group (had received endoscopy more than 3 years before diagnosis), and the unchecked group (had not received endoscopy before diagnosis). The differences in GC staging and treatment modalities among the three groups were analyzed, and the differences in the clinical and pathological features of EGC were further analyzed.

RESULTS

One thousand and twenty-five GC patients were included, with 395 cases of EGC and 630 cases of advanced GC. The proportions of EGC in the short-interval, long-interval, and unchecked groups were 98.0%, 84.2%, and 29.8%, respectively (p < 0.001). Among the 387 lesions of 367 EGC patients were resected by endoscopic submucosal dissection (ESD), 341 (88.1%) exhibited curative resection, and 46 (11.9%) involved noncurative resections. Lesions of EGC differed significantly in diameter, depth of invasion, and curative resection rate (p = 0.033, 0.019, and 0.005, respectively). In the short-interval group, 87.8% of the lesions were ≤ 2 cm, 95.6% of the invasion depths were confined to the mucosal layer, and 96.7% of the eCura scores were A or B. Compared with the unchecked group, they had smaller diameters (RR = 0.419, 95% CI 0.234-0.752), shallower invasion depths (RR = 0.286, 95% CI 0.105-0.777), and a higher curative resection rate (RR = 0.215, 95% CI 0.068-0.676).

CONCLUSION

Endoscopic surveillance at 3-year intervals can help detect EGC, and the EGC lesions found have smaller diameters and shallower depths of invasion, helping improve the curative resection rate of ESD.

Applicability of the Barcelona scale to assess the quality of cleanliness of mucosa at esophagogastroduodenoscopy

BACKGROUND AND OBJECTIVES

There are few scales with prospective validation for the assessment of the upper gastrointestinal mucosal cleanliness during an esophagogastroduodenoscopy (EGD). The aim of this study was to develop a valid and reproducible cleanliness scale for use during an EGD.

METHODS

We developed a cleanliness scale (Barcelona scale) with a score (0-2 points) of five segments of the upper gastrointestinal tract with thorough cleaning techniques (esophagus, fundus, body, antrum, and duodenum). First, 125 photos (25 of each area) were assessed, and a score was assigned to each image by consensus among 7 experts endoscopists. Subsequently, 100 of the 125 images were selected and the inter- and intra-observer variability of 15 previously trained endoscopists was evaluated using the same images at two different times.

RESULTS

In total, 1500 assessments were performed. In 1336/1500 observations (89%) there was agreement with the consensus score, with a mean kappa value of 0.83 (0.45-0.96). In the second evaluation, in 1330/1500 observations (89%) there was agreement with the consensus score, with a mean kappa value of 0.82 (0.45-0.93). The intra-observer variability was 0.89 (0.76-0.99).

CONCLUSIONS

The Barcelona cleanliness scale is a valid measure and reproducible with minimal training. Its application in clinical practice is a significant step to standardize the quality of the EGD.

Mortality burden and future projections of major risk factors for esophageal cancer in China from 1990 to 2019

OBJECTIVE

This study, based on Global Burden of Disease (GBD) data, aimed to report the long-term trend in mortality rates caused by risk factors for esophageal cancer (EC) in China from 1990 to 2019 and predict the burden of EC mortality caused by these risk factors over the next 15 years.

METHODS

We examined six risk factors that influenced EC mortality rates in China and their respective rankings. Furthermore, we analyzed the number of deaths and crude mortality rates (CMR) caused by these risk factors for both sexes and different age groups. Age-standardized mortality rates (ASMR) and the number of deaths across all age groups were also analyzed. Finally, we utilized the Bayesian Age-Period-Cohort (BAPC) model to predict the trends in ASMR burden caused by these risk factors in the future.

RESULTS

From 1990 to 2019, the percentage changes in ASMR for EC caused by the six risk factors in China were as follows: smoking (- 33.4%), alcohol consumption (- 23.0%), low fruit intake (- 73.6%), low vegetable intake (- 96.0%), high Body Mass Index (BMI) (25.1%), and tobacco chewing (- 32.8%). In 2019, the top three risk factors contributing to EC ASMR in China were smoking, alcohol consumption, and high BMI. Overall, the ASMR for EC in China fluctuated and declined from 1990 to 2019. The most common risk factors for males were smoking and alcohol consumption, while low fruit intake and high BMI were the most common risk factors for females. The impact of these risk factors on EC mortality increased with age, except for the elderly population. BAPC analysis indicated that the influence of these risk factors on ASMR was expected to remain relatively stable in the next 15 years, suggesting a continued significant burden of EC.

CONCLUSION

The projected burden of EC mortality in China was expected to continue increasing steadily over the next 15 years, highlighting the pressing need for disease control measures. To alleviate this burden, targeted prevention and control policies addressing risk factors for EC such as smoking, alcohol consumption, and high BMI are necessary.

The Effect of Using Premedication of Simethicone/Pronase With or Without Postural Change on Visualization of the Mucosa Before Endoscopy: A Prospective, Double Blinded, Randomized Controlled Trial

INTRODUCTION

To investigate the efficacy and safety of preprocedural simethicone (S) and pronase (P) for optimal mucosal visualization during esophagogastroduodenoscopy with sedation. The effect of postural change combined with premedication on mucosal visibility was also examined.

METHODS

The study randomized 496 patients into 8 groups based on the type of premedication provided and whether a postural change occurred. The premedication in the control group was 100 mL of normal saline solution (NS). The remaining 3 intervention groups were administered 100 mL of simethicone alone (S), pronase solution alone (P), and simethicone plus pronase solution (S + P). Each group was classified into subgroups according to whether there was a postural change (PC). The mucosal visibility score (MVS), total mucosal visibility score (TVS), procedure time, water consumption for mucosal cleansing, and proportion of patients with diminutive lesions <5 mm were recorded.

RESULTS

The P and S groups had a significantly better TVS than the NS group (11.86 ± 3.36 in group P vs 14.52 ± 2.57 in group NS, P < 0.001; 12.36 ± 2.93 in group S vs 14.52 ± 2.57 in group NS, P = 0.006). The TVS was better in the P group than in the S group (11.86 ± 3.36 vs 12.36 ± 2.93, P = 0.037). The MVS was significantly better in the esophagus and duodenum and worse in the upper and lower gastric body in the S group than in the P group. The P + S group had a significantly better TVS than the P and S groups (9.81 ± 2.90 in group P + S vs 11.86 ± 3.36 in group P and 12.36 ± 2.93 in group S, respectively, P < 0.001),\ and had a reduced amount of flushing water during the procedure (0 [interquartile range [IQR]: 0-33] mL in group P + S vs 40 [IQR: 0-70] mL in group P, P < 0.01; 0 [IQR: 0-33] mL in group P + S vs 50 [IQR: 20-98] mL in group S, P < 0.001). The TVS was significantly better in the P + S + PC group than in the P + S group (8.44 ± 2.10 vs 9.81 ± 2.90, P = 0.003). The MVS was significantly better in the gastric antrum, fundus, and upper and lower gastric body in the P + S + PC group than in the P + S group. There was no significant difference in the detection rate of diminutive lesions among the different groups during an endoscopic examination ( P > 0.05).

DISCUSSION

The combination of preprocedural administration with simethicone and pronase achieved superior mucosal visualization compared with saline, simethicone, or pronase alone in patients receiving upper endoscopy. Postural change maneuvers performed before endoscopy further improved the mucosal visibility in most parts of the stomach when used with preprocedural simethicone and pronase.

Curative criteria for endoscopic treatment of gastric cancer

Endoscopic treatment, particularly endoscopic submucosal dissection, has become the primary treatment for early gastric cancer. A comprehensive optical assessment, including white light endoscopy, image-enhanced endoscopy, and magnification, are the cornerstones for clinical staging and determining the resectability of lesions. This paper discusses factors that influence the indication for endoscopic resection and the likelihood of achieving a curative resection. Our review stresses the critical need for interpreting the histopathological report in accordance with clinical guidelines and the imperative of tailoring decisions based on the patients' and lesions' characteristics and preferences. Moreover, we offer guidance on managing complex scenarios, such as those involving non-curative resection. Finally, we identify future research avenues, including the role of artificial intelligence in estimating the depth of invasion and the urgent need to refine predictive scores for lymph node metastasis and metachronous lesions.

Latest Advances in Endoscopic Detection of Oesophageal and Gastric Neoplasia

Endoscopy is the gold standard for the diagnosis of cancers and cancer precursors in the oesophagus and stomach. Early detection of upper GI cancers requires high-quality endoscopy and awareness of the subtle features these lesions carry. Endoscopists performing surveillance of high-risk patients including those with Barrett's oesophagus, previous squamous neoplasia or chronic atrophic gastritis should be familiar with endoscopic features, classification systems and sampling techniques to maximise the detection of early cancer. In this article, we review the current approach to diagnosis of these conditions and the latest advanced imaging and diagnostic techniques.

Deep Learning and Gastric Cancer: Systematic Review of AI-Assisted Endoscopy

BACKGROUND

Gastric cancer (GC), a significant health burden worldwide, is typically diagnosed in the advanced stages due to its non-specific symptoms and complex morphological features. Deep learning (DL) has shown potential for improving and standardizing early GC detection. This systematic review aims to evaluate the current status of DL in pre-malignant, early-stage, and gastric neoplasia analysis.

METHODS

A comprehensive literature search was conducted in PubMed/MEDLINE for original studies implementing DL algorithms for gastric neoplasia detection using endoscopic images. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The focus was on studies providing quantitative diagnostic performance measures and those comparing AI performance with human endoscopists.

RESULTS

Our review encompasses 42 studies that utilize a variety of DL techniques. The findings demonstrate the utility of DL in GC classification, detection, tumor invasion depth assessment, cancer margin delineation, lesion segmentation, and detection of early-stage and pre-malignant lesions. Notably, DL models frequently matched or outperformed human endoscopists in diagnostic accuracy. However, heterogeneity in DL algorithms, imaging techniques, and study designs precluded a definitive conclusion about the best algorithmic approach.

CONCLUSIONS

The promise of artificial intelligence in improving and standardizing gastric neoplasia detection, diagnosis, and segmentation is significant. This review is limited by predominantly single-center studies and undisclosed datasets used in AI training, impacting generalizability and demographic representation. Further, retrospective algorithm training may not reflect actual clinical performance, and a lack of model details hinders replication efforts. More research is needed to substantiate these findings, including larger-scale multi-center studies, prospective clinical trials, and comprehensive technical reporting of DL algorithms and datasets, particularly regarding the heterogeneity in DL algorithms and study designs.

The role of artificial intelligence in the endoscopic diagnosis of esophageal cancer: a systematic review and meta-analysis

Early detection of esophageal cancer is limited by accurate endoscopic diagnosis of subtle macroscopic lesions. Endoscopic interpretation is subject to expertise, diagnostic skill, and thus human error. Artificial intelligence (AI) in endoscopy is increasingly bridging this gap. This systematic review and meta-analysis consolidate the evidence on the use of AI in the endoscopic diagnosis of esophageal cancer. The systematic review was carried out using Pubmed, MEDLINE and Ovid EMBASE databases and articles on the role of AI in the endoscopic diagnosis of esophageal cancer management were included. A meta-analysis was also performed. Fourteen studies (1590 patients) assessed the use of AI in endoscopic diagnosis of esophageal squamous cell carcinoma-the pooled sensitivity and specificity were 91.2% (84.3-95.2%) and 80% (64.3-89.9%). Nine studies (478 patients) assessed AI capabilities of diagnosing esophageal adenocarcinoma with the pooled sensitivity and specificity of 93.1% (86.8-96.4) and 86.9% (81.7-90.7). The remaining studies formed the qualitative summary. AI technology, as an adjunct to endoscopy, can assist in accurate, early detection of esophageal malignancy. It has shown superior results to endoscopists alone in identifying early cancer and assessing depth of tumor invasion, with the added benefit of not requiring a specialized skill set. Despite promising results, the application in real-time endoscopy is limited, and further multicenter trials are required to accurately assess its use in routine practice.

Editorial: PPIs and gastric cancer - Are we dealing with a carcinogen? Authors' reply

This article is linked to Gong et al papers. To view these articles, visit https://doi.org/10.1111/apt.17676 and https://doi.org/10.1111/apt.17691

Revolutionizing healthcare by use of artificial intelligence in esophageal carcinoma - a narrative review

Esophageal cancer is a major cause of cancer-related mortality worldwide, with significant regional disparities. Early detection of precursor lesions is essential to improve patient outcomes. Artificial intelligence (AI) techniques, including deep learning and machine learning, have proved to be of assistance to both gastroenterologists and pathologists in the diagnosis and characterization of upper gastrointestinal malignancies by correlating with the histopathology. The primary diagnostic method in gastroenterology is white light endoscopic evaluation, but conventional endoscopy is partially inefficient in detecting esophageal cancer. However, other endoscopic modalities, such as narrow-band imaging, endocytoscopy, and endomicroscopy, have shown improved visualization of mucosal structures and vasculature, which provides a set of baseline data to develop efficient AI-assisted predictive models for quick interpretation. The main challenges in managing esophageal cancer are identifying high-risk patients and the disease's poor prognosis. Thus, AI techniques can play a vital role in improving the early detection and diagnosis of precursor lesions, assisting gastroenterologists in performing targeted biopsies and real-time decisions of endoscopic mucosal resection or endoscopic submucosal dissection. Combining AI techniques and endoscopic modalities can enhance the diagnosis and management of esophageal cancer, improving patient outcomes and reducing cancer-related mortality rates. The aim of this review is to grasp a better understanding of the application of AI in the diagnosis, treatment, and prognosis of esophageal cancer and how computer-aided diagnosis and computer-aided detection can act as vital tools for clinicians in the long run.

Lightweight deep learning model incorporating an attention mechanism and feature fusion for automatic classification of gastric lesions in gastroscopic images

Accurate diagnosis of various lesions in the formation stage of gastric cancer is an important problem for doctors. Automatic diagnosis tools based on deep learning can help doctors improve the accuracy of gastric lesion diagnosis. Most of the existing deep learning-based methods have been used to detect a limited number of lesions in the formation stage of gastric cancer, and the classification accuracy needs to be improved. To this end, this study proposed an attention mechanism feature fusion deep learning model with only 14 million (M) parameters. Based on that model, the automatic classification of a wide range of lesions covering the stage of gastric cancer formation was investigated, including non-neoplasm(including gastritis and intestinal metaplasia), low-grade intraepithelial neoplasia, and early gastric cancer (including high-grade intraepithelial neoplasia and early gastric cancer). 4455 magnification endoscopy with narrow-band imaging(ME-NBI) images from 1188 patients were collected to train and test the proposed method. The results of the test dataset showed that compared with the advanced gastric lesions classification method with the best performance (overall accuracy = 94.3%, parameters = 23.9 M), the proposed method achieved both higher overall accuracy and a relatively lightweight model (overall accuracy =95.6%, parameter = 14 M). The accuracy, sensitivity, and specificity of low-grade intraepithelial neoplasia were 94.5%, 93.0%, and 96.5%, respectively, achieving state-of-the-art classification performance. In conclusion, our method has demonstrated its potential in diagnosing various lesions at the stage of gastric cancer formation.

Umbrella systematic review of potential quality indicators for the detection of dysplasia and cancer at upper gastrointestinal endoscopy

Background and study aims Upper gastrointestinal (UGI) endoscopy lacks established quality indicators. We conducted an umbrella systematic review of potential quality indicators for the detection of UGI cancer and dysplasia. Methods Bibliographic databases were searched up to December 2021 for systematic reviews and primary studies. Studies reporting diagnostic accuracy, detection rates or the association of endoscopy or endoscopist-related factors with UGI cancer or dysplasia detection were included. AMSTAR2 and JBI checklists were used to assess systematic review and primary study quality. Clinical heterogeneity precluded meta-analysis and findings are summarized narratively. Results Eight systematic reviews and nine primary studies were included. Image enhancement, especially narrow band imaging, had high diagnostic accuracy for dysplasia and early gastric cancer (pooled sensitivity 0.87 (95% CI 0.84-0.89) and specificity 0.97 (0.97-0.98)). Higher detection rates with longer endoscopy examination times were reported in three studies, but no difference was observed in one study. Endoscopist biopsy rate was associated with increased gastric cancer detection (odds ratio 2.5; 95% confidence interval [CI] 2.1-2.9). Early esophageal cancer (0.17% vs 0.14%, P =0.04) and gastric cancer (0.16% vs 0.12%, P =0.02) detection rates were higher with propofol sedation compared to no sedation. Endoscopies performed by trained endoscopists on dedicated Barrett's surveillance lists had higher detection rates (8% vs 3%, P <0.001). The neoplasia detection rate during diagnostic endoscopies for Barrett's esophagus was 7% (95% CI 4%-10%). Conclusions Image enhancement use, longer examination times, biopsy rate and propofol sedation are potential quality indicators for UGI endoscopy. Neoplasia detection rate and dedicated endoscopy lists are additional potential quality indicators for Barrett's esophagus.

Post-Endoscopy Upper Gastrointestinal Cancer Rate in a Tertiary Referral Centre: An Australian Data Linkage Analysis

OBJECTIVE

High-quality gastroscopy is critical for early diagnosis of upper gastrointestinal cancers (UGCs), and assessment of missed cancers may serve as a key quality metric. Using a prospective gastroscopy database and data linkage with the Queensland Cancer Registry, we assessed the risk of developing UGC within 3 years of a cancer-negative gastroscopy at an Australian tertiary centre. Additional aims were to identify factors predictive of missed cancer, perform root cause analyses for missed cancers and assess overall survival.

DESIGN/METHOD

We identified patients who were diagnosed with UGC within 3 years of undergoing gastroscopy between 2011 and 2016. Non-mucosal cancers, cancers distal to duodenum and patients undergoing surveillance were excluded. Cases diagnosed within 6 months of gastroscopy were defined as detected cancers, while those developing within 6-36 months were defined as missed cancers. Post-endoscopy UGC rate (PEUGIC-3Y) was calculated as ratio of missed over total cancers detected. Demographic, clinical, endoscopic and histologic variables were analysed.

RESULTS

A total of 17,131 gastroscopies were performed for 10,393 patients during the study period. One hundred and twenty-six UGCs were diagnosed, including 120 detected UGCs and 6 missed UGCs. The overall PEUGIC-3Y rate was 4.8% (95% CI 2.1-10.4). The missed UGCs included 3 gastric adenocarcinomas, 2 gastro-oesophageal junction adenocarcinomas and 1 oesophageal squamous cell carcinoma. At the preceding 'cancer-negative gastroscopy', no macroscopic abnormalities were detected at the site of future UGC in 5/6 patients. A UGC developed in 2/6 patients despite an apparent adequate examination at index gastroscopy. Age, sex, indication for endoscopy and cancer location or histology were not predictive of missed cases, and survival was comparable between groups.

CONCLUSION

We demonstrate that the PEUGIC-3Y rate was 4.8% (95% CI 2.1-10.4). The majority of missed cases were adenocarcinomas of the gastro-oesophageal junction or stomach and developed in segments which were found to be normal at index gastroscopy, highlighting the challenges in detecting subtle mucosal lesions in the upper gastrointestinal tract. While overall survival between patients with detected and post-gastroscopy cancers was comparable, these ultimately represent potential missed opportunities for diagnosing an early cancer and underscore the need for quality improvement in gastroscopy.

Comparing artificial intelligence to humans for endoscopic diagnosis of gastric neoplasia: An external validation study

OBJECTIVES

Artificial intelligence (AI) uses deep learning functionalities that may enhance the detection of early gastric cancer during endoscopy. An AI-based endoscopic system for upper endoscopy was recently developed in Japan. We aim to validate this AI-based system in a Singaporean cohort.

METHODS

There were 300 de-identified still images prepared from endoscopy video files obtained from subjects that underwent gastroscopy in National University Hospital (NUH). Five specialists and 6 non-specialists (trainees) from NUH were assigned to read and categorize the images into "neoplastic" or "non-neoplastic." Results were then compared with the readings performed by the endoscopic AI system.

RESULTS

The mean accuracy, sensitivity, and specificity for the 11 endoscopists were 0.847, 0.525, and 0.872, respectively. These values for the AI-based system were 0.777, 0.591, and 0.791, respectively. While AI in general did not perform better than endoscopists on the whole, in the subgroup of high-grade dysplastic lesions, only 29.1% were picked up by the endoscopist rating, but 80% were classified as neoplastic by AI (P = 0.0011). The average diagnostic time was also faster in AI compared with endoscopists (677.1 s vs 42.02 s (P < 0.001).

CONCLUSION

We demonstrated that an AI system developed in another health system was comparable in diagnostic accuracy in the evaluation of static images. AI systems are faster and not fatigable and may have a role in augmenting human diagnosis during endoscopy. With more advances in AI and larger studies to support its efficacy it would likely play a larger role in screening endoscopy in future.

Gastric Intestinal Metaplasia: Challenges and the Opportunity for Precision Prevention

GIM is a persistent, premalignant lesion whereby gastric mucosa is replaced by metaplastic mucosa resembling intestinal tissue, arising in the setting of chronic inflammation, particularly in the context of Helicobacter pylori. While the overall rates of progression to gastric adenocarcinoma are low, estimated at from 0.25 to 2.5%, there are features that confer a much higher risk and warrant follow-up. In this review, we collate and summarise the current knowledge regarding the pathogenesis of GIM, and the clinical, endoscopic and histologic risk factors for cancer. We examine the current state-of-practice with regard to the diagnosis and management of GIM, which varies widely in the published guidelines and in practice. We consider the emerging evidence in population studies, artificial intelligence and molecular markers, which will guide future models of care. The ultimate goal is to increase the detection of early gastric dysplasia/neoplasia that can be cured while avoiding unnecessary surveillance in very low-risk individuals.

Computer-aided demarcation of early gastric cancer: a pilot comparative study with endoscopists

BACKGROUND

Precise area diagnosis of early gastric cancer (EGC) is critical for reliable endoscopic resection. Computer-aided diagnosis (CAD) shows strong potential for detecting EGC and reducing cancer-care disparities caused by differences in endoscopists' skills. To be used in clinical practice, CAD should enable both the detection and the demarcation of lesions. This study proposes a scheme for the detection and delineation of EGC under white-light endoscopy and validates its performance using 1-year consecutive cases.

METHODS

Only 300 endoscopic images randomly selected from 68 consecutive cases were used for training a convolutional neural network. All cases were treated with endoscopic submucosal dissection, enabling the accumulation of a training dataset in which the extent of lesions was precisely determined. For validation, 462 cancer images and 396 normal images from 137 consecutive cases were used. From the validation results, 38 randomly selected images were compared with those delineated by six endoscopists.

RESULTS

Successful detections of EGC in 387 cancer images (83.8%) and the absence of lesions in 307 normal images (77.5%) were achieved. Positive and negative predictive values were 81.3% and 80.4%, respectively. Successful detection was achieved in 130 cases (94.9%). We achieved precise demarcation of EGC with a mean intersection over union of 66.5%, showing the extent of lesions with a smooth boundary; the results were comparable to those achieved by specialists.

CONCLUSIONS

Our scheme, validated using 1-year consecutive cases, shows potential for demarcating EGC. Its performance matched that of specialists; it might therefore be suitable for clinical use in the future.

Setting 6-Minute Minimal Examination Time Improves the Detection of Focal Upper Gastrointestinal Tract Lesions During Endoscopy: A Multicenter Prospective Study

INTRODUCTION

Positive correlation between examination time and neoplasm detection using esophagogastroduodenoscopy (EGD) has been described by observational studies, but the effect of setting minimal examination time still requires investigation.

METHODS

This prospective, 2-stage, interventional study was conducted in 7 tertiary hospitals in China, enrolling consecutive patients undergoing intravenously sedated diagnostic EGDs. In stage I, the baseline examination time was collected without informing the endoscopists. In stage II, the minimal examination time was set for the same endoscopist according to the median examination time of normal EGDs in stage I. The primary outcome was the focal lesion detection rate (FDR), defined as the proportion of subjects with at least one focal lesion among all subjects.

RESULTS

A total of 847 and 1,079 EGDs performed by 21 endoscopists were included in stages I and II, respectively. In stage II, the minimal examination time was set as 6 minutes, and the median time for normal EGD increased from 5.8 to 6.3 minutes ( P < 0.001). Between the 2 stages, the FDR was significantly improved (33.6% vs 39.3%, P = 0.011), and the effect of the intervention was significant (odds ratio, 1.25; 95% confidence interval, 1.03-1.52; P = 0.022) even after adjusting for subjects' age, smoking status, endoscopists' baseline examination time, and working experience. The detection rate of high-risk lesions (neoplastic lesions and advanced atrophic gastritis) was also significantly higher in stage II (3.3% vs 5.4%, P = 0.029). In the endoscopist-level analysis, all practitioners reached a median examination time of 6 minutes, and the coefficients of variation of FDR (36.9%-26.2%) and examination time (19.6%-6.9%) decreased in stage II.

DISCUSSION

Setting a 6-minute minimal examination time significantly improved the detection of focal lesions during EGDs and has the potential to be implemented for quality improvement.

Clinical features of gastric adenoma detected within 3 years after negative screening endoscopy in Korea

Background

Early detection and management of gastric adenoma are important for preventing gastric cancer. The present study aimed to evaluate the predictors of missed gastric adenoma on screening endoscopy in Korea and identify the risk factors associated with interval precancerous gastric lesions.

Methods

All cases of gastric adenomas diagnosed via screening endoscopy between 2007 and 2019 were reviewed. Among them, those who had undergone endoscopy within 3 years were included in the present study. Missed gastric adenoma was defined as gastric adenoma diagnosed within 3 years after negative screening endoscopy.

Results

In total, 295 cases of gastric adenoma were identified. Of these, 95 (32.2%) were missed gastric adenoma cases (mean age, 60.6 years; average interval between final and index endoscopies, 12.6 months); the remaining 200 (67.8%) were newly detected adenoma cases. Univariate analysis revealed that male sex, endoscopist experience, observation time, and presence of gastric intestinal metaplasia (pathologically proven) were associated with missed gastric adenoma. Multivariate analysis revealed that gastric intestinal metaplasia (odds ratio [OR], 2.736; 95% confidence interval [CI], 1.320-5.667; P = 0.007) and shorter observation time of the index screening endoscopy (B, -0.011; OR, 0.990; 95% CI, 0.986-0.993; P < 0.001) were independent risk factors for missed gastric adenoma. The optimal cut-off for the observation time for detecting gastric adenoma was 3.53 minutes (area under curve, 0.738; 95% CI, 0.677-0.799; P < 0.001).

Conclusions

Gastric intestinal metaplasia is an indication of missed gastric adenoma. Therefore, careful inspection of gastric mucosa with gastric intestinal metaplasia and proper observation time can lower the possibility of missing the gastric adenoma during screening.

Variation in the rate of detection of minute and small early gastric cancers at diagnostic endoscopy may reflect the performance of individual endoscopists

OBJECTIVE

The documented variation in gastric cancer (GC) detection among endoscopists has often been dismissed as a coincidental artefact of the low incidence of gastric neoplasms; it is not considered associated with differences in physicians' performance of the esophagogastroduodenoscopy procedure. This study is to confirm whether significant variations among endoscopists in early GC detection suggest the individual performance of the upper endoscopy.

DESIGN

A retrospective observational study at a single centre in Japan assessed the results of 218 early GCs detected during 25 688 routine esophagogastroduodenoscopies by 12 endoscopists. The main outcome was the rate of early GC detection for each endoscopist under the same circumstances. Other measures included the major diameters and locations of the lesions, Helicobacter pylori infection status, and baseline patient characteristics that could affect the prevalence of GC.

RESULTS

The early GC detection rates exhibited wide variation among endoscopists (0.09%-2.87%) despite performing routine esophagogastroduodenoscopies in a population with a similar background. Endoscopists were assigned to a low-detection group (n=6; detection rate: 0.47% (range: 0.09%-0.55%)) and a high-detection group (n=5; detection rate: 0.83% (range: 0.63%-1.12%)), with the single highest detector analysed separately due to his distinct detection rate (2.87%). Endoscopists in the high-detection group had better detection rates for minute (major diameter ≤5 mm) and small (major diameter 6-10 mm) GCs than the low-detection group (0.19%/0.23% vs 0.085%/0.098%). These differences were significant (p<0.01), although there were no significant differences in detection of larger tumours (major diameter ≥11 mm; 0.40% vs 0.28%; p=0.13). The tumour location and H. pylori status were similar in the low-detection group, high-detection group and for the highest detector.

CONCLUSION

Significant variation in the detection of hard-to-find, smaller GCs may reflect individual performance of the examination.

Toward next-generation endoscopes integrating biomimetic video systems, nonlinear optical microscopy, and deep learning

According to the World Health Organization, the proportion of the world's population over 60 years will approximately double by 2050. This progressive increase in the elderly population will lead to a dramatic growth of age-related diseases, resulting in tremendous pressure on the sustainability of healthcare systems globally. In this context, finding more efficient ways to address cancers, a set of diseases whose incidence is correlated with age, is of utmost importance. Prevention of cancers to decrease morbidity relies on the identification of precursor lesions before the onset of the disease, or at least diagnosis at an early stage. In this article, after briefly discussing some of the most prominent endoscopic approaches for gastric cancer diagnostics, we review relevant progress in three emerging technologies that have significant potential to play pivotal roles in next-generation endoscopy systems: biomimetic vision (with special focus on compound eye cameras), non-linear optical microscopies, and Deep Learning. Such systems are urgently needed to enhance the three major steps required for the successful diagnostics of gastrointestinal cancers: detection, characterization, and confirmation of suspicious lesions. In the final part, we discuss challenges that lie en route to translating these technologies to next-generation endoscopes that could enhance gastrointestinal imaging, and depict a possible configuration of a system capable of (i) biomimetic endoscopic vision enabling easier detection of lesions, (ii) label-free in vivo tissue characterization, and (iii) intelligently automated gastrointestinal cancer diagnostic.

Artificial intelligence in theranostics of gastric cancer, a review

Gastric cancer (GC) is one of the commonest cancers with high morbidity and mortality in the world. How to realize precise diagnosis and therapy of GC owns great clinical requirement. In recent years, artificial intelligence (AI) has been actively explored to apply to early diagnosis and treatment and prognosis of gastric carcinoma. Herein, we review recent advance of AI in early screening, diagnosis, therapy and prognosis of stomach carcinoma. Especially AI combined with breath screening early GC system improved 97.4 % of early GC diagnosis ratio, AI model on stomach cancer diagnosis system of saliva biomarkers obtained an overall accuracy of 97.18 %, specificity of 97.44 %, and sensitivity of 96.88 %. We also discuss concept, issues, approaches and challenges of AI applied in stomach cancer. This review provides a comprehensive view and roadmap for readers working in this field, with the aim of pushing application of AI in theranostics of stomach cancer to increase the early discovery ratio and curative ratio of GC patients.

Metachronous Multiple Gastric Cancer Discovered as Endoscopic Curability C2 during Regular Follow-Up after Gastric Endoscopic Submucosal Dissection

INTRODUCTION

The objective of this study was to clarify characteristics of metachronous endoscopic curability C2 (eCura C2) cancer during post-endoscopic submucosal dissection (ESD) follow-up.

METHODS

Of 4,355 gastric lesions treated by ESD at our hospital during 2005-2021, 657 were metachronous. After excluding lesions found ≥2 years since the prior examination or in the gastric remnant, the remaining 515 were analyzed. Study 1: We compared 35 eCura C2 cancers and 480 eCura A-C1 cancers. Study 2: Endoscopic findings of the 35 lesions were examined to determine why they had been missed.

RESULTS

Mean tumor size was larger (34.0 mm vs. 12.1 mm, p < 0.01) and the proportions of mixed-type and poorly differentiated cancers were higher (highly:mixed:poorly, 34.3:57.1:8.6 vs. 94.2:5.0:0.8, p < 0.01) in the eCura C2 group. Study 2: At the prior examination, 4 lesions were noticed but considered benign, 2 lacked sufficient imaging, 19 were detectable on imaging but missed, and 10 were not detectable on imaging. Over half the lesions that were detectable but missed at the prior examination were in the lesser curvature, many being type IIa-IIb lesions with color similar to the background mucosa. All lesions not detectable on imaging at the prior examination were mixed-type or poorly differentiated type.

DISCUSSION

Metachronous cancer detected as eCura C2 cancers was significantly larger, and a significantly higher proportion was mixed-type or poorly differentiated cancers, compared with eCura A-C1 cancers. Possible reasons why these lesions were missed include rapid progression of mixed-type and poorly differentiated cancers, and poor recognition that lesions showing only slight color changes may be present at the lesser curvature.

Orientation in upper gastrointestinal endoscopy—the only way is up

BACKGROUND

Oesophagogastroduodenoscopy is the gold standard investigation for the upper gastrointestinal (UGI) tract. Orientation during endoscopy is challenging and United Kingdom training focusses on technical competence and procedural safety. The reported location of UGI pathologies is crucial to post-endoscopic planning.

AIM

To evaluate endoscopists’ ability to spatially orientate themselves within the UGI tract.

METHODS

A cross sectional descriptive study elicited, using an anonymised survey, the ability of endoscopists to orientate themselves within the UGI tract. The primary outcome was percentage of correct answers from all surveyed; secondary outcomes were percentage of correct answers from experienced vs novice endoscopists. Pearson’s χ² test was applied to compare groups.

RESULTS

Of 188 respondents, 86 were experienced endoscopists having completed over 1000 endoscopies. 44.4% of respondents correctly identified the anterior stomach and 47.3% correctly identified the posterior of the second part of the duodenum (D2). Experienced endoscopists were significantly more likely than novice to identify the anterior stomach correctly [61.6% vs 31.3%, X² (1, n = 188) = 11.10, P = 0.001]. There was no significant difference between the two groups in identifying the posterior of D2.

CONCLUSION

The majority of endoscopists surveyed were unable to identify key landmarks within the UGI tract. Endoscopic orientation appears to improve with experience yet there are some areas still not well recognised. This has potential considerable impact on post-endoscopic management of patients with posterior duodenal ulcers being more likely to perforate and associated with a higher rebleeding risk. We suggest the development of a consensus statement on endoscopic description.