Diagnosing and grading gastric atrophy and intestinal metaplasia using semi-supervised deep learning on pathological images: development and validation study.

OBJECTIVE: Patients with gastric atrophy and intestinal metaplasia (IM) were at risk for gastric cancer, necessitating an accurate risk assessment. We aimed to establish and validate a diagnostic approach for gastric biopsy specimens using deep learning and OLGA/OLGIM for individual gastric cancer risk classification. METHODS: In this study, we prospectively enrolled 545 patients suspected of atrophic gastritis during endoscopy from 13 tertiary hospitals between December 22, 2017, to September 25, 2020, with a total of 2725 whole-slide images (WSIs). Patients were randomly divided into a training set (n = 349), an internal validation set (n = 87), and an external validation set (n = 109). Sixty patients from the external validation set were randomly selected and divided into two groups for an observer study, one with the assistance of algorithm results and the other without. We proposed a semi-supervised deep learning algorithm to diagnose and grade IM and atrophy, and we compared it with the assessments of 10 pathologists. The model's performance was evaluated based on the area under the curve (AUC), sensitivity, specificity, and weighted kappa value. RESULTS: The algorithm, named GasMIL, was established and demonstrated encouraging performance in diagnosing IM (AUC 0.884, 95% CI 0.862-0.902) and atrophy (AUC 0.877, 95% CI 0.855-0.897) in the external test set. In the observer study, GasMIL achieved an 80% sensitivity, 85% specificity, a weighted kappa value of 0.61, and an AUC of 0.953, surpassing the performance of all ten pathologists in diagnosing atrophy. Among the 10 pathologists, GasMIL's AUC ranked second in OLGA (0.729, 95% CI 0.625-0.833) and fifth in OLGIM (0.792, 95% CI 0.688-0.896). With the assistance of GasMIL, pathologists demonstrated improved AUC (p = 0.013), sensitivity (p = 0.014), and weighted kappa (p = 0.016) in diagnosing IM, and improved specificity (p = 0.007) in diagnosing atrophy compared to pathologists working alone. CONCLUSION: GasMIL shows the best overall performance in diagnosing IM and atrophy when compared to pathologists, significantly enhancing their diagnostic capabilities.

The role of the endoscopic grading of gastric intestinal metaplasia in assessing gastric cancer risk: A systematic review and meta-analysis.

Background and aim: Patients with gastric intestinal metaplasia (IM) are at increased risk of gastric cancer (GC). The endoscopic grading of gastric intestinal metaplasia (EGGIM) with high-definition endoscopes has shown the potential to facilitate GC risk stratification. However, a comprehensive review and meta-analysis of published articles are lacking. We conducted a meta-analysis to access the value of EGGIM in the assessment of histological IM. Materials: Studies were selected from PubMed, Medline, Embase, and Cochrane (last selection, Jun 2022). We extracted relevant data to calculate the accuracy of EGGIM compared with the operative link of gastric intestinal metaplasia (OLGIM) and to calculate pooled odds ratio (OR) with a 95% confidence interval (CI) assessing GC risk with different grading. Results: Four diagnostic studies and three case-control clinical trials were included in the analysis, which included 665 patients and 738 patients, respectively. Compared with OLGIM III/IV, EGGIM(5-10) had a pooled sensitivity and specificity of 0.92(95%CI 0.86-0.96) and 0.90(95%CI 0.88-0.93), and the area under the curve(AUC) was 0.9702. In assessing early GC, the pooled OR of patients with EGGIM(5-10) was 7.46(95%CI 3.41-16.31) compared with that of EGGIM(0-4). Conclusions: EGGIM is highly consistent with OLGIM, and patients with EGGIM(5-10) are at a higher risk for early GC. Some heterogeneity in the current research suggests that we need to carry out more strict control of confounding factors. Systematic Review Registration: [https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=248691], (Prospero registration number: 248691).

Efficacy and safety of berberine in preventing recurrence of colorectal adenomas: A systematic review and meta-analysis.

ETHNOPHARMACOLOGICAL RELEVANCE: Berberine(BBR) is a kind of isoquinoline alkaloids extracted from the rhizomes of Coptis chinensis Franch., which was the main active ingredient. Accumulating evidence has shown that it has potential pharmacological effects in preventing the recurrence of colorectal adenomas. AIM OF THE STUDY: The roles of BBR in the overall recurrence of colorectal adenoma have still not been assessed because of the limitations of the available data and the restriction of a single study. Therefore, we evaluated the effectiveness and safety of BBR in preventing the recurrence of colorectal adenomas through a systematic review and meta-analysis of available data. MATERIALS AND METHODS: We searched four English databases (PubMed (MEDLINE), the Cochrane Central Register of Controlled Trials (CENTRAL), Embase and Web of Science) and four Chinese language databases (Chinese Biomedicine (CBM), China National Knowledge Infrastructure (CNKI), Chinese Scientific Journals Database (VIP) and the WanFang Database) from their inception through October 2020. Meta-analysis was performed with RevMan5.3 software after data extraction and the quality of studies assessment. RESULTS: Three randomized controlled clinical trials were included with 1076 patients. Our results illustrated that 1-year and 2-year supplementation with BBR was associated with lower recurrence rate of colorectal adenoma (RR 0.69, 95% CI 0.57 to 0.84, p=0.0001; RR 0.75, 95% CI 0.64 to 0.88, p=0.0004). The relative risk of oral BBR for 1 year and 2 years is not comparable, for 2-year efficacy outcomes were assessed in all participants who had at least one colonoscopy with pathological evaluation after baseline (lots of participants completed the first colonoscopy but discontinued during the second follow-up interval.). Moreover, the results also suggest that BBR had more adverse events than placebo (RR 2.91, 95% CI 1.24 to 6.85, p=0.01). Through the full-text reading, no serious adverse events were observed, and constipation was the most common event which disappears once the drug is discontinued. CONCLUSION: Generally, the present study indicated that BBR has a comparable therapeutic effect on the prevention of colorectal adenomas recurrence. Adverse reactions are worthy of attention which requires additional studies to obtain a precise conclusion. PROSPERO REGISTRATION NO: CRD42020209135.

Gastric Cancer Risk of Intestinal Metaplasia Subtypes: A Systematic Review and Meta-Analysis of Cohort Studies.

INTRODUCTION: Intestinal metaplasia (IM) is an independent risk factor for gastric cancer (GC). However, the subtypes of IM as a risk factor for GC remain controversial. We performed a systematic review and meta-analysis to evaluate the relationship between IM subtypes and GC risk. METHODS: Systematic searches were conducted in PubMed, EMBASE, and the Cochrane Library for published cohort studies of patients with complete IM (type I) or incomplete IM (type II or type III) from inception to May 15, 2021. We extracted relevant data and calculated pooled risk ratios (RRs) and 95% confidence intervals (CIs) comparing the GC risk with IM subtypes. RESULTS: Twelve cohort studies comprising 6,498 individuals were included in the study. Compared with complete IM, the pooled relative risk of GC risk of patients with incomplete IM was 5.16 (95% CI, 3.28-8.12), and the GC risk of type III IM was the highest, with a pooled relative risk of 2.88 (95% CI, 1.37-6.04) compared with that of type II. Compared with complete IM, the pooled relative risk of dysplasia risk in patients with incomplete IM was 3.72 (95% CI, 1.42-9.72), and the dysplasia risk of type III IM was 11.73 (95% CI, 2.08-66.08) compared with that of type I. DISCUSSION: Patients with incomplete IM, especially type III, were at a higher risk of GC and dysplasia than those with complete IM. The current evidence indicates a potential correlation between IM subtypes and GC risk, which may support the use of IM subtypes in GC surveillance.

The prognostic role of a combined fibrinogen and inflammation-based index in patients with metastatic breast cancer.

BACKGROUND: The activation of inflammation and coagulation cascades plays an essential role in the development of various malignancies, including metastatic breast cancer (MBC). This retrospective study aimed to investigate the prognostic role of the combination of fibrinogen and the inflammation-based index in patients with MBC. METHODS: A total of 176 patients with MBC were retrospectively reviewed. The clinical and pathological data of included patients were followed-up and analyzed. The plasma fibrinogen concentration (FIB), neutrophil-lymphocyte ratio (NLR), and platelet-lymphocyte ratio (PLR) were measured. Dynamic variations in the FIB, NLR, and PLR values were collected from 56 MBC patients before and after first-line therapy. Receiver operating characteristic (ROC) curves were constructed to assess the optimal cut-off values. Correlations between FIB and NLR or PLR were evaluated using Spearman correlation analysis. The Kaplan-Meier method, two-tailed log-rank test, and Cox proportional hazard model were used for statistical analysis. RESULTS: Baseline FIB was positively correlated with NLR and PLR values in MBC patients (P<0.05). Additionally, multivariable analysis proved that the ERBB2 + subtype (P=0.023), basal-like subtype (P=0.032), targeted therapy (P=0.033), other regimens (P=0.005), and baseline FIB level (P=0.004) were independent prognostic variables for progression-free survival (PFS) in MBC patients. Furthermore, ERBB2+, basal-like subtypes, and baseline hyperfibrinogenemia were independent factors for poor prognosis in MBC patients [hazard ratio (HR): 3.717, 95% confidence interval (CI): 1.561-8.851, P=0.003; HR: 3.245, 95% CI: 1.368-7.698, P=0.008; HR: 2.069, 95% CI: 1.352-3.167, P=0.001, respectively]. Most importantly, the FIB level increased significantly after first-line therapy in patients with disease progression (3.73±0.63 vs. 5.32±0.52 g/L, P=0.042) and also decreased markedly in stable disease (3.42±1.05 vs. 3.03±0.73 g/L, P=0.036). However, PFS and overall survival (OS) were not significantly correlated with the dynamic changes of FIB and the inflammation-based index. CONCLUSIONS: The present study provided evidence that baseline FIB combined with NLR and PLR could serve as prognostic predictors for MBC patients. Dynamic change of FIB before and after first-line therapy could also be used as a potential predictor of therapeutic response in MBC patients.

Using neural networks to estimate the losses of ascorbic acid, total phenols, flavonoid, and antioxidant activity in asparagus during thermal treatments.

Artificial neural networks (ANNs) with back-propagation algorithm were developed to predict the percentage loss of ascorbic acid, total phenols, flavonoid, and antioxidant activity in different segments of asparagus during water blanching at temperatures ranging from 65 to 95 degrees C as a function of blanching time and temperature. In this study, the one-hidden-layer ANNs are used, and the number of neurons in the hidden layer were chosen by trial and error. Optimized ANN models were developed for predicting nutrient losses in bud, upper, middle, and butt segments of asparagus. ANN models were then tested against an independent data set. Our results showed that the predicted values of the correlation coefficients between experimental and ANNs ranged from 0.8166 to 0.9868. Therefore, ANNs could be potential tools for the prediction of nutrient losses in vegetables during thermal treatments.