[Application of machine learning model based on XGBoost algorithm in early prediction of patients with acute severe pancreatitis].

OBJECTIVE: To establish a machine learning model based on extreme gradient boosting (XGBoost) algorithm for early prediction of severe acute pancreatitis (SAP), and explore its predictive efficiency. METHODS: A retrospective cohort study was conducted. The patients with acute pancreatitis (AP) who admitted to the First Affiliated Hospital of Soochow University, the Second Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University from January 1, 2020 to December 31, 2021 were enrolled. Demography information, etiology, past history, and clinical indicators and imaging data within 48 hours of admission were collected according to the medical record system and image system, and the modified CT severity index (MCTSI), Ranson score, bedside index for severity in acute pancreatitis (BISAP) and acute pancreatitis risk score (SABP) were calculated. The data sets of the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University were randomly divided into training set and validation set according to 8 : 2. Based on XGBoost algorithm, the SAP prediction model was constructed on the basis of hyperparameter adjustment by 5-fold cross validation and loss function. The data set of the Second Affiliated Hospital of Soochow University was served as independent test set. The predictive efficacy of the XGBoost model was evaluated by drawing the receiver operator characteristic curve (ROC curve), and compared it with the traditional AP related severity score; variable importance ranking diagram and Shapley additive explanation (SHAP) diagram were drawn to visually explain the model. RESULTS: A total of 1 183 AP patients were enrolled finally, of which 129 (10.9%) developed SAP. Among the patients from the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University, there were 786 patients in the training set and 197 in the validation set; 200 patients from the Second Affiliated Hospital of Soochow University were used as the test set. Analysis of all three datasets showed that patients who advanced to SAP exhibited pathological manifestation such as abnormal respiratory function, coagulation function, liver and kidney function, and lipid metabolism. Based on the XGBoost algorithm, an SAP prediction model was constructed, and ROC curve analysis showed that the accuracy for prediction of SAP reached 0.830, the area under the ROC curve (AUC) was 0.927, which was significantly improved compared with the traditional scoring systems including MCTSI, Ranson, BISAP and SABP, the accuracy was 0.610, 0.690, 0.763, 0.625, and the AUC was 0.689, 0.631, 0.875, and 0.770, respectively. The feature importance analysis based on the XGBoost model showed that the top ten items ranked by the importance of model features were admission pleural effusion (0.119), albumin (Alb, 0.049), triglycerides (TG, 0.036), Ca2+ (0.034), prothrombin time (PT, 0.031), systemic inflammatory response syndrome (SIRS, 0.031), C-reactive protein (CRP, 0.031), platelet count (PLT, 0.030), lactate dehydrogenase (LDH, 0.029), and alkaline phosphatase (ALP, 0.028). The above indicators were of great significance for the XGBoost model to predict SAP. The SHAP contribution analysis based on the XGBoost model showed that the risk of SAP increased significantly when patients had pleural effusion and decreased Alb. CONCLUSIONS: A SAP prediction scoring system was established based on the machine automatic learning XGBoost algorithm, which can predict the SAP risk of patients within 48 hours of admission with good accuracy.

Helicobacter pylori infection related long noncoding RNA (lncRNA) AF147447 inhibits gastric cancer proliferation and invasion by targeting MUC2 and up-regulating miR-34c.

Long non-coding RNAs (lncRNAs) were shown to play critical roles in cancer biology. We investigated whether H. pylori infection could promote gastric cancer by regulating lncRNAs expression. Differentially expressed lncRNAs between H. pylori positive and negative tissues were identified by microarray and validated by qRT-PCR. Our results indicated that H. pylori positive tissues have a specific profile of lncRNAs. Cell biological assays with siRNA-mediated knockdown or lentivirus vector-mediated over-expression were performed to probe the functional relevance of the lncRNAs. We identified an lncRNA-AF147447 decreased expressed by H. pylori infection, which can inhibit GC proliferation and invasion in vitro and in vivo, act as a tumor suppressor in the development of H. pylori induced GC. LncRNA AF147447 could repress MUC2 expression by direct binding or increasing miR-34c expression. We also found that transcription factor E2F1 could be recruited to lncRNA AF147447 promoter by RNA immunoprecipatation and RNA pull down assays. These findings support a role of lncRNA AF147447 in tumor suppression. This discovery contributes to a better understanding of the importance of the deregulated lncRNAs by H. pylori infection and provides a rationale for the potential development of lncRNA-based targeted approaches for the treatment of H. pylori-related gastric cancer.

Incretin-Based Therapy and Risk of Pancreatic Cancer in Patients with Type 2 Diabetes Mellitus: A Meta-analysis of Randomized Controlled Trials.

INTRODUCTION: The present study aims to evaluate the risk of pancreatic cancer with incretin-based therapy among patients with type 2 diabetes mellitus (T2DM). METHODS: We searched EMBASE, MEDLINE, the Cochrane Central Register of Controlled Trials and ClinicalTrials.gov for eligible studies published up to March 06 2016. This meta-analysis includes all studies reporting adverse events of pancreatic cancer with use of incretin-based therapies compared with placebo or non-incretin anti-diabetic drugs in patients with T2DM. We used fixed-effect model to compare pooled relative risk (RR) with related 95% confidence intervals (CI). RESULTS: A total of 159 randomized trials were identified. Out of these, 135 studies were excluded as pancreatic cancer occurrence had not been included as an end point. The remaining 24 trials enrolling 47,904 participants were further assessed. Overall, no increased risk of pancreatic cancer were detected in association with incretin-based treatment (RR = 0.7, 95% CI 0.37-1.05). The incidence of pancreatic neoplasm was even lower among incretin-based groups than controls (RR = 0.50, 95% CI 0.29-0.87) in trials with duration more than 104 weeks. There was even decreased risk of pancreatic cancer within groups paralleled by incretin-matched placebos (RR = 0.55, 95% CI 0.32-0.93) than by non-incretin anti-diabetic drugs. Neither monotherapy (RR = 0.62, 95% CI 0.38-1.01) nor combination regimen (RR = 0.92, 95% CI 0.45-1.90) of incretin mimetics increased the risk of pancreatic cancer. CONCLUSION: This meta-analysis shows that incretin-based therapies are not associated with increase in the risk of pancreatic cancer. Interestingly, subgroup analyses suggested lower risk of pancreatic cancer in incretin groups than placebo in long-term studies (>104 weeks). Considering the inconsistent results among randomized trials and previous epidemiological investigations, more such studies should be conducted to clarify the existence or non-existence of this association. FUNDING: This work was supported by grants from the National Natural Science Foundation of China (Nos. 81270476 and 81470830).

MiR-141 Inhibits Gastric Cancer Proliferation by Interacting with Long Noncoding RNA MEG3 and Down-Regulating E2F3 Expression.

BACKGROUND: MiR-141 and long noncoding RNA MEG3 have been independently reported to be tumor suppressor genes in various cancers. However, their expression has never been previously associated with gastric cancer (GC). AIMS: To investigate the interaction of miR-141 and MEG3 in GC. METHODS: QRT-PCR was used to detect miR-141, MEG3, and E2F3 in gastric tissues and cells. CCK-8 and flow cytometry analysis were used to detect cell functions. Western blot and luciferase activity were used to identify E2F3 as one of the direct targets of miR-141. RESULTS: We found that expression of both miR-141 and MEG3 was significantly reduced in GC compared with levels in matched nonmalignant tissues. Positive correlation between miR-141 and MEG3 was found in both tumor tissues and control tissues. Furthermore, the over-expression of either miR-141 or MEG3 in 7901 and MKN45 cells inhibited cell proliferation and cell cycle progression and promoted cell apoptosis. E2F3 was identified as a target of miR-141, and its inhibition significantly reduced MEG3 expression. E2F3 expression was also found to be negatively associated with both MEG3 and miR-141. E2F3 over-expression partly reversed the changes caused by transfection of miR-141 mimic, and inhibition of miR-141 or MEG3 overrides MEG3- or miR-141-induced modulation of cell growth in GC. CONCLUSIONS: These findings together suggested that miR-141 could be interacting with MEG3 and targeting E2F3, and these factors may play important anti-tumor effects in GC pathogenesis and provide therapeutic targets in the clinics.

Helicobacter pylori Infection Is Associated with an Increased Risk of Hyperemesis Gravidarum: A Meta-Analysis.

Background. Several studies have shown a possible involvement of Helicobacter pylori (H. pylori) infection in individuals with hyperemesis gravidarum (HG), but the relationship remains controversial. This meta-analysis was performed to validate and strengthen the association between HG and H. pylori infection. Methods. PubMed, Embase, and Web of Science databases up to March 20, 2014, were searched to select studies on the prevalence of H. pylori infection between pregnant women with HG and the normal pregnant control subjects. Results. Of the HG cases, 1289 (69.6%) were H. pylori-positive; however, 1045 (46.2%) were H. pylori-positive in control group. Compared to the non-HG normal pregnant controls, infection rate of H. pylori was significantly higher in pregnant women with HG (OR = 3.34, 95% CI: 2.32-4.81, P < 0.001). Subgroup analysis indicated that H. pylori infection was a risk factor of HG in Asia, Africa, and Oceania, especially in Africa (OR = 12.38, 95% CI: 7.12-21.54, P < 0.001). Conclusions. H. pylori should be considered one of the risk factors of HG, especially in the developing countries. H. pylori eradication could be considered to relieve the symptoms of HG in some intractable cases.

Association between Helicobacter pylori Infection and Chronic Urticaria: A Meta-Analysis.

Background. Some studies have shown the possible involvement of Helicobacter pylori (H. pylori) infection in chronic urticaria, but the relationship remains controversial. The aim of this meta-analysis was to quantitatively assess the association between H. pylori infection and chronic urticaria. Methods. Observational studies comparing the prevalence of H. pylori infection in patients with chronic urticaria and control subjects were identified through a systematic search in MEDLINE and EMBASE up to July 2014. H. pylori infection was confirmed by serological or nonserological tests. For subgroup analyses, studies were separated by region, publication year, and H. pylori detection method to screen the potential factors resulting in heterogeneity. Results. 16 studies involving 965 CU cases and 1235 controls were included. Overall, the prevalence of H. pylori infection was higher in urticarial patients than in controls (OR = 1.66; 95% CI: 1.12-2.45; P = 0.01). This result persisted in subanalysis of nine high-quality studies (OR = 1.36; 95% CI: 1.03-1.80; P = 0.03). Subgroup analysis showed that detection method of H. pylori is also a potential influential factor for the overall results. Conclusions. Our present meta-analysis suggests that H. pylori infection is significantly, though weakly, associated with an increased risk of chronic urticaria.

Association between recurrent aphthous stomatitis and Helicobacter pylori infection: a meta-analysis.

OBJECTIVES: Several studies have shown the possible involvement of Helicobacter pylori infection in individuals with recurrent aphthous stomatitis (RAS), but the relationship remains controversial. This meta-analysis was performed to validate the association between RAS and H. pylori infection. MATERIALS AND METHODS: The PubMed database was searched up to January 25, 2013 to select studies on the prevalence of H. pylori infection between RAS patients and control subjects. Studies were included if they evaluated and clearly defined exposure to RAS, reported the incidence of H. pylori infection, or provided data for their estimation. For subgroup analyses, studies were separated by region, publication year, and source of controls to screen the potential factors against the results. Before meta-analysis, the studies were evaluated for publication bias and heterogeneity. Summary odds ratio (OR) estimates with 95 % confidence intervals (CIs) were calculated using the fixed-effects model. RESULTS: Seven case-control studies containing 339 cases and 271 controls were eventually selected for analysis. A total of 100 (29.50 %) RAS patients had H. pylori infection, which was significantly greater than the 54 (19.93 %) non-RAS controls with H. pylori infection (OR = 1.85, 95 % CI: 1.24-2.74, P = 0.002). This result persisted in a hospital-based control subgroup (OR = 2.72, 95 % CI: 1.57-4.72). CONCLUSIONS: Based on our meta-analysis, H. pylori infection is associated with an increased risk of RAS. CLINICAL RELEVANCE: The eradication of H. pylori in the stomach may promote relief of RAS symptoms and healing of oral ulcers, and even prevent the occurrence of RAS.