ObjectiveTo explore the mechanism by which exosomes derived from gastric cancer cells MGC-803 promote tumor progression through the upregulation of ABCG8 expression. MethodsUltra-performance liquid chromatography-tandem mass spectrometry （UPLC-MS/MS） lipidomics， quantitative real-time PCR （qPCR） and Western blot were used to verify the role of ABCG8 in gastric cancer progression. Exosomes derived from gastric cancer cells were introduced， and a series of in vitro experiments were conducted to evaluate their effect on ABCG8 expression， as well as on cell proliferation， migration， and clonogenic ability in normal gastric mucosal epithelial GES-1 cells. ResultsUPLC-MS/MS lipidomics analysis revealed significant changes in triacylglycerol （TG） among the differential lipids. qPCR and Western blot analysis showed that the mRNA and protein expression levels of ABCG8 in the GC-Exo-treated group were higher than those in the GES-1 cell group （P<0.05）. Colony formation and Transwell migration assays revealed that， compared with the GES-1 cell group， the GC-Exo-treated group exhibited a greater number of colonies （P<0.05） and migrated cells （P<0.05） after 48 h of treatment. CCK-8 assay results showed that the proliferation rate in the GC-Exo-treated group （50 μg/mL） was higher than that in the GES-1 cell group （P<0.001）. Lipidomics analysis revealed that， compared with the GES-1 cell group， TG levels were upregulated in the GC-Exo-treated group， accompanied by decreased levels of phosphatidylcholine （PC） PC（38：2e）. ConclusionExosomes derived from gastric cancer cells MGC-803 may promote tumor progression by upregulating ABCG8 expression to mediate cholesterol efflux， thereby triggering pro-carcinogenic lipid metabolic reprogramming in normal gastric mucosal epithelial cells.

Objective:To evaluate the efficacy and safety of endoscopic injection sclerotherapy (EIS) and endoscopic rubber band ligation (ERBL) for the treatment of grade Ⅰ-Ⅲ internal hemorrhoids.Methods:A prospective study was conducted on patients with grade Ⅰ-Ⅲ internal hemorrhoids who sought treatment at Renmin Hospital of Wuhan University or the NO.2 People's Hospital of Fuyang City from November 2021 to November 2022. Eligible patients were continuously enrolled based on inclusion and exclusion criteria and randomized into 2 groups using a central randomization system: the EIS group and the ERBL group. The primary outcomes included symptom improvement rate, recurrence rate, incidence of adverse events (bleeding, anal distension, pain, urinary retention, etc.), surgical costs, patient satisfaction, hemorrhoidal disease symptom score (HDSS), European quality of life 5-dimensions (EQ-5D) score, and self-rated health status score.Results:A total of 203 patients were enrolled (86 from Renmin Hospital of Wuhan University and 117 from the NO.2 People's Hospital of Fuyang City), with 103 in the EIS group and 100 in the ERBL group. Both groups successfully completed endoscopic treatment for internal hemorrhoids. The surgical cost in the EIS group was significantly lower than that in the ERBL group (1 044.77±522.77 yuan VS 2 538.44±465.63 yuan, t=-21.660, P<0.001). The incidence of perioperative pain and moderate-to-severe pain in the EIS group was significantly lower than that in the ERBL group [2.91% (3/103) VS 25.00% (25/100), χ2=20.817, P<0.001; 0.97% (1/103) VS 18.00% (18/100), χ2=17.344, P<0.001]. There were no significant differences in the incidence of perioperative bleeding, anal distension, or urinary retention between the two groups [0.97% (1/103) VS 1.00% (1/100), 11.65% (12/103) VS 19.00% (19/100), 0.00% (0/103) VS 2.00% (2/100), P>0.05]. During the 12-week follow-up, 4 patients were lost to follow-up (all from the EIS group). There were no significant differences in symptom improvement rate, recurrence rate, or patient satisfaction rate between the two groups [96.97% (96/99) VS 96.00% (96/100), 3.03% (3/99) VS 5.00% (5/100), 97.98% (97/99) VS 95.00% (95/100), P>0.05]. At 12 weeks postoperatively, the HDSS in the EIS group significantly decreased compared to preoperative levels [0.0 (0.0, 1.0) VS 5.0 (3.0, 7.0), Z=-18.270, P<0.010], the EQ-5D score in the ERBL group significantly increased compared to preoperative levels (1.00±0.01 VS 0.98±0.03, F=27.527, P<0.010), and self-rated health status score in the ERBL group significantly increased compared to preoperative levels (92.31±6.89 VS 82.62±10.98, F=115.025, P<0.010). At 12 weeks postoperatively, the HDSS in the ERBL group significantly decreased compared to preoperative levels [0.0 (0.0, 1.0) VS 5.0 (4.0, 8.0), Z=-16.110, P<0.010], the EQ-5D score in the ERBL group significantly increased compared to preoperative levels (1.00±0.00 VS 0.98±0.05, F=13.718, P<0.010), and self-rated health status score in the ERBL group significantly increased compared to preoperative levels (93.46±6.35 VS 84.15±10.71, F=123.695, P<0.010). Conclusion:Both EIS and ERBL are safe and effective treatments for grade Ⅰ-Ⅲ internal hemorrhoids, with high patient satisfaction. Comparatively, EIS demonstrates lower surgical costs and a reduced incidence and severity of perioperative pain.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.

Objective:To construct and validate an artificial intelligence diagnostic system based on multi-feature integration for diagnosing gastric whitish neoplastic lesions under white-light endoscopy.Methods:Gastroscopic images from Renmin Hospital of Wuhan University and the Seventh Medical Center of Chinese PLA General Hospital were collected from November 2012 to July 2021. A total of 823 images of gastric whitish lesions from 267 patients were finally selected. Five white-light endoscopic features associated with gastric whitish lesions were selected through a literature search, including lesion location, boundary clarity, surface texture, roundness, and depression status. Images with manually annotated features were used to train machine learning models, with the optimal model selected as the multi-feature fitting diagnostic system, which assigned diagnostic weights to each feature. A conventional deep learning model was trained with the same dataset. The diagnostic performance of the two models were compared, and eight endoscopists of varying expertise were invited to participate in human-machine comparisons.Results:Accuracy, sensitivity, and specificity of the multi-feature fitting diagnostic system were 82.11% (101/123), 78.43% (40/51), and 84.72% (61/72), respectively. Feature weights in descending order were depression (0.71), lesion location (0.11), surface roughness (0.08), boundary clarity (0.06), and subcircular shape (0.04). The diagnostic accuracy of the system was significantly higher than that of non-expert endoscopists (82.11% VS 74.31%, Z=-2.785, P=0.008) and comparable to that of expert endoscopists (82.11% VS 83.20%, Z=-0.696, P=0.700). There was no significant difference in accuracy between the multi-feature fitting diagnostic system and the traditional deep learning model [82.11% (101/123) VS 82.93% (102/123), P=1.000]. Conclusion:The feature-weighted artificial intelligence diagnostic system for gastric whitish neoplastic lesions demonstrates clinically relevant diagnostic accuracy under white-light endoscopy.

Objective:To analyze interobserver variability in endoscopic diagnostic accuracy of chronic atrophic gastritis (CAG) among endoscopists with varying levels of experience.Methods:Endoscopic examination data from 247 patients who underwent endoscopy from January 2021 to June 2024 at Department of Gastroenterology, the First Affiliated Hospital of Army Medical University ( n=154), Renmin Hospital of Wuhan University ( n=35) and Shanxi Provincial Coal Central Hospital ( n=58) were retrospectively collected. The collected images were reviewed by an expert panel of three individuals with the title of deputy chief physician or above. The final endoscopic Kimura-Takemoto classification diagnosis of the expert panel was regarded as the golden standard. Fourteen endoscopists from the above three centers provided their Kimura-Takemoto classification diagnosis. These endoscopists were divided into the junior group ( n=7, with experience of <2 000 procedures) and the senior group ( n=7, with experience of >10 000 procedures). The difference in the accuracy of endoscopic Kimura-Takemoto classification diagnosis between the groups were analyzed. Results:Diagnostic accuracy for Kimura-Takemoto classification ranged from 65.99% (163/247) to 86.64% (214/247) in the senior group with the overall accuracy of 77.27% (1 336/1 729). The junior group exhibited diagnostic accuracy ranging from 36.44% (90/247) to 72.47% (179/247) with the overall accuracy of 62.12% (1 074/1 729). The senior group demonstrated higher overall diagnostic accuracy than that of the junior group ( χ 2=93.27, P<0.001). The diagnostic accuracy of non-CAG in the senior group was higher than that in the junior group [83.73% (463/553) VS 72.33% (400/553), χ 2=20.27, P<0.001]. The diagnostic accuracy of C-type atrophy in the senior group was higher than that in the junior group [90.10% (801/889) VS 82.79% (736/889), χ 2=19.66, P<0.001] .The diagnostic accuracy of O-type atrophy in the senior group was higher than that in the junior group [83.97% (241/287) VS 68.29% (196/287), χ 2=18.56, P<0.001]. Conclusion:Interobserver variability is observed in the diagnostic accuracy of endoscopic Kimura-Takemoto classification for CAG among endoscopists with different experience levels. Experienced endoscopists exhibit higher diagnostic accuracy for CAG compared with their less experienced counterparts.

Objective:To analyze interobserver variability in endoscopic diagnostic accuracy of chronic atrophic gastritis (CAG) among endoscopists with varying levels of experience.Methods:Endoscopic examination data from 247 patients who underwent endoscopy from January 2021 to June 2024 at Department of Gastroenterology, the First Affiliated Hospital of Army Medical University ( n=154), Renmin Hospital of Wuhan University ( n=35) and Shanxi Provincial Coal Central Hospital ( n=58) were retrospectively collected. The collected images were reviewed by an expert panel of three individuals with the title of deputy chief physician or above. The final endoscopic Kimura-Takemoto classification diagnosis of the expert panel was regarded as the golden standard. Fourteen endoscopists from the above three centers provided their Kimura-Takemoto classification diagnosis. These endoscopists were divided into the junior group ( n=7, with experience of <2 000 procedures) and the senior group ( n=7, with experience of >10 000 procedures). The difference in the accuracy of endoscopic Kimura-Takemoto classification diagnosis between the groups were analyzed. Results:Diagnostic accuracy for Kimura-Takemoto classification ranged from 65.99% (163/247) to 86.64% (214/247) in the senior group with the overall accuracy of 77.27% (1 336/1 729). The junior group exhibited diagnostic accuracy ranging from 36.44% (90/247) to 72.47% (179/247) with the overall accuracy of 62.12% (1 074/1 729). The senior group demonstrated higher overall diagnostic accuracy than that of the junior group ( χ 2=93.27, P<0.001). The diagnostic accuracy of non-CAG in the senior group was higher than that in the junior group [83.73% (463/553) VS 72.33% (400/553), χ 2=20.27, P<0.001]. The diagnostic accuracy of C-type atrophy in the senior group was higher than that in the junior group [90.10% (801/889) VS 82.79% (736/889), χ 2=19.66, P<0.001] .The diagnostic accuracy of O-type atrophy in the senior group was higher than that in the junior group [83.97% (241/287) VS 68.29% (196/287), χ 2=18.56, P<0.001]. Conclusion:Interobserver variability is observed in the diagnostic accuracy of endoscopic Kimura-Takemoto classification for CAG among endoscopists with different experience levels. Experienced endoscopists exhibit higher diagnostic accuracy for CAG compared with their less experienced counterparts.

Objective:To evaluate the efficacy and safety of endoscopic injection sclerotherapy (EIS) and endoscopic rubber band ligation (ERBL) for the treatment of grade Ⅰ-Ⅲ internal hemorrhoids.Methods:A prospective study was conducted on patients with grade Ⅰ-Ⅲ internal hemorrhoids who sought treatment at Renmin Hospital of Wuhan University or the NO.2 People's Hospital of Fuyang City from November 2021 to November 2022. Eligible patients were continuously enrolled based on inclusion and exclusion criteria and randomized into 2 groups using a central randomization system: the EIS group and the ERBL group. The primary outcomes included symptom improvement rate, recurrence rate, incidence of adverse events (bleeding, anal distension, pain, urinary retention, etc.), surgical costs, patient satisfaction, hemorrhoidal disease symptom score (HDSS), European quality of life 5-dimensions (EQ-5D) score, and self-rated health status score.Results:A total of 203 patients were enrolled (86 from Renmin Hospital of Wuhan University and 117 from the NO.2 People's Hospital of Fuyang City), with 103 in the EIS group and 100 in the ERBL group. Both groups successfully completed endoscopic treatment for internal hemorrhoids. The surgical cost in the EIS group was significantly lower than that in the ERBL group (1 044.77±522.77 yuan VS 2 538.44±465.63 yuan, t=-21.660, P<0.001). The incidence of perioperative pain and moderate-to-severe pain in the EIS group was significantly lower than that in the ERBL group [2.91% (3/103) VS 25.00% (25/100), χ2=20.817, P<0.001; 0.97% (1/103) VS 18.00% (18/100), χ2=17.344, P<0.001]. There were no significant differences in the incidence of perioperative bleeding, anal distension, or urinary retention between the two groups [0.97% (1/103) VS 1.00% (1/100), 11.65% (12/103) VS 19.00% (19/100), 0.00% (0/103) VS 2.00% (2/100), P>0.05]. During the 12-week follow-up, 4 patients were lost to follow-up (all from the EIS group). There were no significant differences in symptom improvement rate, recurrence rate, or patient satisfaction rate between the two groups [96.97% (96/99) VS 96.00% (96/100), 3.03% (3/99) VS 5.00% (5/100), 97.98% (97/99) VS 95.00% (95/100), P>0.05]. At 12 weeks postoperatively, the HDSS in the EIS group significantly decreased compared to preoperative levels [0.0 (0.0, 1.0) VS 5.0 (3.0, 7.0), Z=-18.270, P<0.010], the EQ-5D score in the ERBL group significantly increased compared to preoperative levels (1.00±0.01 VS 0.98±0.03, F=27.527, P<0.010), and self-rated health status score in the ERBL group significantly increased compared to preoperative levels (92.31±6.89 VS 82.62±10.98, F=115.025, P<0.010). At 12 weeks postoperatively, the HDSS in the ERBL group significantly decreased compared to preoperative levels [0.0 (0.0, 1.0) VS 5.0 (4.0, 8.0), Z=-16.110, P<0.010], the EQ-5D score in the ERBL group significantly increased compared to preoperative levels (1.00±0.00 VS 0.98±0.05, F=13.718, P<0.010), and self-rated health status score in the ERBL group significantly increased compared to preoperative levels (93.46±6.35 VS 84.15±10.71, F=123.695, P<0.010). Conclusion:Both EIS and ERBL are safe and effective treatments for grade Ⅰ-Ⅲ internal hemorrhoids, with high patient satisfaction. Comparatively, EIS demonstrates lower surgical costs and a reduced incidence and severity of perioperative pain.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.

Objective:To construct and validate an artificial intelligence diagnostic system based on multi-feature integration for diagnosing gastric whitish neoplastic lesions under white-light endoscopy.Methods:Gastroscopic images from Renmin Hospital of Wuhan University and the Seventh Medical Center of Chinese PLA General Hospital were collected from November 2012 to July 2021. A total of 823 images of gastric whitish lesions from 267 patients were finally selected. Five white-light endoscopic features associated with gastric whitish lesions were selected through a literature search, including lesion location, boundary clarity, surface texture, roundness, and depression status. Images with manually annotated features were used to train machine learning models, with the optimal model selected as the multi-feature fitting diagnostic system, which assigned diagnostic weights to each feature. A conventional deep learning model was trained with the same dataset. The diagnostic performance of the two models were compared, and eight endoscopists of varying expertise were invited to participate in human-machine comparisons.Results:Accuracy, sensitivity, and specificity of the multi-feature fitting diagnostic system were 82.11% (101/123), 78.43% (40/51), and 84.72% (61/72), respectively. Feature weights in descending order were depression (0.71), lesion location (0.11), surface roughness (0.08), boundary clarity (0.06), and subcircular shape (0.04). The diagnostic accuracy of the system was significantly higher than that of non-expert endoscopists (82.11% VS 74.31%, Z=-2.785, P=0.008) and comparable to that of expert endoscopists (82.11% VS 83.20%, Z=-0.696, P=0.700). There was no significant difference in accuracy between the multi-feature fitting diagnostic system and the traditional deep learning model [82.11% (101/123) VS 82.93% (102/123), P=1.000]. Conclusion:The feature-weighted artificial intelligence diagnostic system for gastric whitish neoplastic lesions demonstrates clinically relevant diagnostic accuracy under white-light endoscopy.

Objective To investigate the effect of different timing of endoscopic intervention on the short-term prognosis of patients with cirrhosis combined with esophageal and gastric variceal bleeding(EGVB),as well as to analyze the independent related factors affecting 6-week mortality of patients with EGVB and to construct a prediction model.Methods The patients with EGVB admitted to Renmin Hospital of Wuhan University for emergency endoscopic intervention from June 2021 to December 2023 were consecutively included.According to the timing of endoscopic intervention,they were categorized into the urgent endoscopy group(within 6 hours after the consultation,n=94)and the early endoscopy group(6-24 hours after the consultation,n=76).All the enrolled patients underwent a 6-week postoperative follow-up.The propensity score matching(PSM)was used as 1∶1,and baseline characteristics before and after PSM,operation related information,short-term prognosis,etc.after PSM between the two groups were compared.The binary logistic regression was used to analyze the factors influencing the 6-week mortality of patients with EGVB.A nomogram model based on the independent related factors were constructed,and ROC,calibration,and decision curves were used to verify the prognostic value of the model.Results There were significant differences in the model of end-stage liver disease(MELD)score,total bilirubin level,propotions of patients with systolic blood pressure＜90 mmHg,diabetes mellitus or initial bleeding event between the two groups(P＜0.05).After PSM,a total of 106 patients met the criteria.There was no significant difference in baseline date,operation related information,complications after operation,and 6-week mortality rate after operation between the two groups with 53 patients in each group after PSM.Blood product consumption was significantly higher in the urgent endoscopy group than that in the early endoscopy group(P=0.046).The binary logistic regression analysis results showed that initial bleeding,higher Charson comorbidity index(CCI)score,active bleeding and longer duration of surgery were independent related factors for 6-week mortality after operation in patients with EGVB(P＜0.05),whereas timing of endoscopic intervention was not a related factor for 6-week mortality.A nomogram model was constructed and AUC of the model was 0.790,with higher accuracy and utility.Conclusions The association of endoscopic intervention timing with 6-week mortality after operation in patients with EGVB is not significant obvious;the prognostic value of the model basing on initial bleeding,CCI score,active bleeding and surgery duration is higher for 6-week mortality after operation in patients with EGVB.