Objective To investigate the effect of transcutaneous auricular vagus nerve stimulation(taVNS)on quality of early postoperative recovery in patients undergoing thoracoscopic lung resection.Methods A total of 168 patients scheduled for elective thoracoscopic lung resection(wedge resection,segmentectomy,lobectomy)under general anesthesia were enrolled and randomly assigned to active-taVNS group(group T)or sham-taVNS group(group S)(n=84).Participants received four consecutive 30-minute sessions of active stimulation or sham stimulation at four time points:(1)the afternoon prior to the surgery,(2)the morning of the surgery,(3)following extubation,and(4)the first afternoon post-surgery.The Quality of Recovery-15(QoR-15)scores of the patients,the Numerical Rating Scale(NRS)scores at rest and during cough at 24,48,and 72 hours after surgery were recorded;and the usage of opioids within 48 hours after surgery was recorded;the duration of chest tube indwelling,incidence of severe pulmonary complications,postoperative hospital stay and adverse reactions to the stimulation(such as nausea and vomiting,fever,constipation,dizziness and itching)were observed.Results Compared with group S,group T exhibited significantly higher QoR-15 scores at 24,48,and 72 h postoperatively,lower NRS pain scores during resting and coughing,and reduced opioid consumption within 48 hours postoperatively(P<0.05).There were no significant differences between the two groups in the duration of chest tube indwelling,incidence of severe pulmonary complications,hospital stay,and the incidence of adverse reactions to the stimula-tion(P>0.05).Conclusion TaVNS can significantly improve quality of early postoperative recovery in patients undergoing thoracoscopic lung resection,and provide more effective postoperative analgesia without increasing the risk of postoperative complications.

Objective To explore the short-term prognosis and risk factors for in-hospital mortality in patients with fulminant myocarditis induced refractory cardiogenic shock(FM-RCS)receiving veno-arterial extracorporeal membrane oxygenation(VA-ECMO)treatment,and to construct an early prognosis prediction model using relevant indicators.Methods A total of 61 FM-RCS patients treatment by VA-ECMO in the department of intensive care unit of the Second Affiliated Hospital of Zhengzhou University from January 2017 to February 2024,excluding 15 cases with age less than 18 years and 3 cases with ECMO treatment duration less than 24 hours,a total of 43 patients were finally included.Participants were stratified into survival(n=19)and mortality(n=24)groups according to discharge outcomes.Demographic data,chronic disease history,early laboratory indicators,left ventricular function indicators,and basic reference values of hemodynamics were systematically compared between the two groups.Variable selection was performed using LASSO regression,followed by multivariate COX regression analysis to screen independent risk factors for in-hospital mortality in ECMO-treatment FM-RCS patients.A nomogram prediction model was subsequently developed using R software and validated through calibration curves,concordance index(C-index),and receiver operator characteristic curve(ROC curve)analysis.Results The overall survival rate of the 43 enrolled patients was 44.2%,with 19 cases in the survival group and 24 cases in the mortality group.In early laboratory indicators,the survival group exhibited significantly lower levels of initial lactic acid(Lac),24-hour Lac(Lac 24 h),24-hour MB isoenzyme of creatine kinase(CK-MB 24 h),24-hour cardiac troponin T(cTnT 24 h),24-hour total bilirubin(TBil 24 h),24-hour serum creatinine(SCr 24 h),and lactate albumin ratio(LAR)compared to the mortality group[initial Lac(mmol/L):2.7(1.3,7.6)vs.9.2(5.9,14.0),Lac 24 h(mmol/L):2.4(2.0,3.6)vs.5.4(3.3,9.2),CK-MB 24 h(U/L):58.0(28.0,115.0)vs.167.7(68.5,280.3),cTnT 24 h(μg/L):0.53(0.37,2.41)vs.3.92(3.10,8.86),TBil 24h(μmol/L):18.3(9.9,37.8)vs.40.2(24.6,67.0),SCr 24 h(μmol/L):90.63±42.49 vs.177.76±70.76,LAR:0.09(0.04,0.23)vs.0.31(0.20,0.38),all P<0.05],serum albumin(Alb)levels were significantly higher in the survival group[g/L:36.0(31.9,39.2)vs.31.7(26.4,34.4),P<0.05].The mortality group had a higher incidence of malignant arrhythmias[66.7%(16/24)vs.31.6%(6/19),P<0.05].The LASSO regression model identified four non-zero coefficient variables-Lac 24 h,CK-MB 24 h,cTnT 24 h,and SCr 24 h-which were included in the subsequent multivariate COX regression analysis.The results demonstrated that Lac 24 h[hazard ratio(HR)and 95%confidence interval(95%CI)was 1.186(1.074-1.310),P<0.001]and cTnT 24 h(HR=1.230,95%CIwas 1.078-1.404,P=0.002)were independent risk factors for in-hospital mortality in VA-ECMO treatment FM-RCS patients.A predictive model constructed using these two indicators showed a C-index of 0.812,area under the curve(AUC)=0.941,with 91.7%sensitivity and 94.7%specificity.Furthermore,compared to the survival group,the mortality group exhibited significantly higher incidences of acute kidney injury[91.7%(22/24)vs.36.8%(7/19)]and hypoxic-ischemic encephalopathy[62.5%(15/24)vs.10.5%(2/19),both P<0.05].The mortality group also required greater transfusion volumes[mL:3 800(1 420,8 515)vs.1 200(400,3 020),P<0.05],but had shorter total hospitalization durations[days:7(3,13)vs.23(20,44),P<0.05].Conclusion For FM-RCS patients receiving VA-ECMO treatment,Lac 24 h and cTnT 24 h after ECMO initiation are independent predictors of in-hospital mortality.Clinicians should be vigilant about poor prognosis in FM-RCS patients with high Lac 24 h hours(>2.5 mmol/L)and cTnT 24 hours(>3.01 μg/L)after ECMO treatment.

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 transcutaneous auricular vagus nerve stimulation(taVNS)on quality of early postoperative recovery in patients undergoing thoracoscopic lung resection.Methods A total of 168 patients scheduled for elective thoracoscopic lung resection(wedge resection,segmentectomy,lobectomy)under general anesthesia were enrolled and randomly assigned to active-taVNS group(group T)or sham-taVNS group(group S)(n=84).Participants received four consecutive 30-minute sessions of active stimulation or sham stimulation at four time points:(1)the afternoon prior to the surgery,(2)the morning of the surgery,(3)following extubation,and(4)the first afternoon post-surgery.The Quality of Recovery-15(QoR-15)scores of the patients,the Numerical Rating Scale(NRS)scores at rest and during cough at 24,48,and 72 hours after surgery were recorded;and the usage of opioids within 48 hours after surgery was recorded;the duration of chest tube indwelling,incidence of severe pulmonary complications,postoperative hospital stay and adverse reactions to the stimulation(such as nausea and vomiting,fever,constipation,dizziness and itching)were observed.Results Compared with group S,group T exhibited significantly higher QoR-15 scores at 24,48,and 72 h postoperatively,lower NRS pain scores during resting and coughing,and reduced opioid consumption within 48 hours postoperatively(P<0.05).There were no significant differences between the two groups in the duration of chest tube indwelling,incidence of severe pulmonary complications,hospital stay,and the incidence of adverse reactions to the stimula-tion(P>0.05).Conclusion TaVNS can significantly improve quality of early postoperative recovery in patients undergoing thoracoscopic lung resection,and provide more effective postoperative analgesia without increasing the risk of postoperative complications.

Objective To explore the short-term prognosis and risk factors for in-hospital mortality in patients with fulminant myocarditis induced refractory cardiogenic shock(FM-RCS)receiving veno-arterial extracorporeal membrane oxygenation(VA-ECMO)treatment,and to construct an early prognosis prediction model using relevant indicators.Methods A total of 61 FM-RCS patients treatment by VA-ECMO in the department of intensive care unit of the Second Affiliated Hospital of Zhengzhou University from January 2017 to February 2024,excluding 15 cases with age less than 18 years and 3 cases with ECMO treatment duration less than 24 hours,a total of 43 patients were finally included.Participants were stratified into survival(n=19)and mortality(n=24)groups according to discharge outcomes.Demographic data,chronic disease history,early laboratory indicators,left ventricular function indicators,and basic reference values of hemodynamics were systematically compared between the two groups.Variable selection was performed using LASSO regression,followed by multivariate COX regression analysis to screen independent risk factors for in-hospital mortality in ECMO-treatment FM-RCS patients.A nomogram prediction model was subsequently developed using R software and validated through calibration curves,concordance index(C-index),and receiver operator characteristic curve(ROC curve)analysis.Results The overall survival rate of the 43 enrolled patients was 44.2%,with 19 cases in the survival group and 24 cases in the mortality group.In early laboratory indicators,the survival group exhibited significantly lower levels of initial lactic acid(Lac),24-hour Lac(Lac 24 h),24-hour MB isoenzyme of creatine kinase(CK-MB 24 h),24-hour cardiac troponin T(cTnT 24 h),24-hour total bilirubin(TBil 24 h),24-hour serum creatinine(SCr 24 h),and lactate albumin ratio(LAR)compared to the mortality group[initial Lac(mmol/L):2.7(1.3,7.6)vs.9.2(5.9,14.0),Lac 24 h(mmol/L):2.4(2.0,3.6)vs.5.4(3.3,9.2),CK-MB 24 h(U/L):58.0(28.0,115.0)vs.167.7(68.5,280.3),cTnT 24 h(μg/L):0.53(0.37,2.41)vs.3.92(3.10,8.86),TBil 24h(μmol/L):18.3(9.9,37.8)vs.40.2(24.6,67.0),SCr 24 h(μmol/L):90.63±42.49 vs.177.76±70.76,LAR:0.09(0.04,0.23)vs.0.31(0.20,0.38),all P<0.05],serum albumin(Alb)levels were significantly higher in the survival group[g/L:36.0(31.9,39.2)vs.31.7(26.4,34.4),P<0.05].The mortality group had a higher incidence of malignant arrhythmias[66.7%(16/24)vs.31.6%(6/19),P<0.05].The LASSO regression model identified four non-zero coefficient variables-Lac 24 h,CK-MB 24 h,cTnT 24 h,and SCr 24 h-which were included in the subsequent multivariate COX regression analysis.The results demonstrated that Lac 24 h[hazard ratio(HR)and 95%confidence interval(95%CI)was 1.186(1.074-1.310),P<0.001]and cTnT 24 h(HR=1.230,95%CIwas 1.078-1.404,P=0.002)were independent risk factors for in-hospital mortality in VA-ECMO treatment FM-RCS patients.A predictive model constructed using these two indicators showed a C-index of 0.812,area under the curve(AUC)=0.941,with 91.7%sensitivity and 94.7%specificity.Furthermore,compared to the survival group,the mortality group exhibited significantly higher incidences of acute kidney injury[91.7%(22/24)vs.36.8%(7/19)]and hypoxic-ischemic encephalopathy[62.5%(15/24)vs.10.5%(2/19),both P<0.05].The mortality group also required greater transfusion volumes[mL:3 800(1 420,8 515)vs.1 200(400,3 020),P<0.05],but had shorter total hospitalization durations[days:7(3,13)vs.23(20,44),P<0.05].Conclusion For FM-RCS patients receiving VA-ECMO treatment,Lac 24 h and cTnT 24 h after ECMO initiation are independent predictors of in-hospital mortality.Clinicians should be vigilant about poor prognosis in FM-RCS patients with high Lac 24 h hours(>2.5 mmol/L)and cTnT 24 hours(>3.01 μg/L)after ECMO treatment.

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.

Intrahepatic cholangiocarcinoma (ICC) is a highly malignant tumor with poor prognosis. In the process of clinical diagnosis and treatment, only a small number of patients can receive surgical treatment in time, and the rest often have local infiltration or distant metastasis at the time of presentation, which can only prolong the overall survival by adjuvant therapy. At present, the main adjuvant treatments in clinical practice include chemotherapy, targeted therapy and immunotherapy. This article reviews the progress of systemic therapy and some clinical trials in patients with advanced unresectable ICC, to provide a reference for the clinical diagnosis and treatment of ICC.

Objective:To compare the feasibility and efficacy of translocator protein (TSPO) molecular probes N, N-diethyl-2-(2-(4- 18F-fluorophenyl)-5, 7-dimethylpyrazolo[1, 5-a]pyrimidin-3-yl)acetamide ( 18F-FDPA) and 18F-(R)-( N-sec-butyl)-3-fluoromethyl- N-methyl-4-phenylquinoline-2-carboxamide (LW223) for the detection of abdominal vulnerable atherosclerotic plaques (VAP) in rabbit models. Methods:Nine healthy New Zealand white rabbits were divided into group A (control group, n=3), group B (VAP group, n=3) and group C (VAP treatment group, n=3) using completely randomized design. Animals were injected with 18F-FDPA and 18F-LW223 at the end of 12, 16 and 24 weeks. PET/CT and CT angiography (CTA) was performed 40-50 min post injection. All rabbits were sacrificed at the end of 24 weeks after imaging studies. All abdominal aortas were collected for pathological and immunofluorescence examination. Repeated measures analysis of variance (Bonferroni test) and paired t-test were used to analyze the data. Results:Target-to-background ratio (TBR; abdominal aortic lesion/left ventricular blood pool) values of 18F-FDPA in 3 groups at the end of 12, 16 and 24 weeks were significantly different ( F values: 68.09-144.88, all P<0.001). At the end of 12 weeks, there was no increased uptake of 18F-FDPA in the abdominal aorta region in 3 groups. The local 18F-FDPA uptake of the abdominal aorta in group B was significantly higher than the uptake in group C and that in group A at the end of 16 and 24 weeks( P<0.05 or P<0.001), and there were significant differences between group C and group A, with higher uptake in group C (both P<0.01). In 3 groups, there was no significant 18F-LW223 uptake in the abdominal aorta region at 3 time points of PET/CTA imaging. At the end of 12, 16 and 24 weeks, TBR values of 18F-FDPA and 18F-LW223 in 3 groups exhibited statistical differences ( t values: 2.88-36.79, all P<0.05). HE, immunofluorescent CD68 and TSPO staining showed more macrophage infiltration in group B than group C. Conclusion:18F-FDPA can be used to detect VAP in rabbits′ abdominal arteries at early time compared to 18F-LW223, and to evaluate the changes in the stability of vulnerable plaque after lipid-lowering drug intervention.