Objective: To investigate the clinical effect of plasma exchange (PE) versus double plasma molecular adsorption system combined with PE (DPMAS+PE) in the treatment of toxic hepatic failure. Methods: A total of 67 patients with toxic hepatic failure who were admitted during the same period of time were divided into PE group, DPMAS+ PE group, and control group. The 22 patients in the PE group were treated with PE alone, and the 24 patients in the DPMAS+PE group were given DPMAS combined with PE. The clinical out-come was compared between the three groups. Results: Both treatment groups had significantly higher clinical response rate and 24-week survival rate than the control group. After treatment, both treatment groups had significant reductions in the serum levels of total bilirubin (TBil) , direct bilirubin (DBil) , alanine aminotransfer-ase (ALT) , and aspartate aminotransferase (AST) , and the PE group had significant reductions in the albumin (Alb) level and activated partial thromboplastin time (APTT) (P<0.05) ; the DPMAS+PE group showed no sig-nificant changes in the Alb level and APTT (P>0.05). There were no significant differences in TBil, DBil, ALT, and AST between the two treatment groups after treatment (P>0.05). After treatment, the PE group had significantly higher Alb level and APTT than the DPMAS + PE group (P<0.05). Compared with the control group, both treatment groups had significant reductions in TBil, DBil, ALT and AST after treatment (P<0.05). Conclusion The two artificial liver support techniques can significantly improve patients' liver function and in-crease their survival rate, and the combined artificial liver support technique can reduce the amount of plasma used.

The transcriptional repressor RE1 silencer transcription factor(NRSF/REST) is an important factor that restricts some neuronal traits in neurons.Since these traits are also present in pancreatic islet cells,NRSF-regulated genes involved in islet function are searched.A NRSE-like motif was analysed in human insulin promoter.The role of NRSE was evaluated by generating a model of insulin-secreting cells that firmly express NRSF.The presence of NRSF led to a decrease in activity of human insulin promoter by stable or transient transfection with human insulin-promoter luciferase.The predicted NRSE-like motif also confers NRSF-dependent transcriptional repression in the context of a surrogate gene promoter.Specific binding activity of NRSF/REST to the NRSE-like motif was confirmed by EMSA.Moreover,the binding activity is competed by consensus NRSE sequence.These data showed that human insulin promoter is regulated by the transcriptional repressor NRSF/REST via the NRSE-like motif.

Antibiotic-associated diarrhea(AAD) is a frequent adverse effect of antibiotic in children.AAD is associated with longer hospitalization, higher healthcare cost and even lead to death.Pediatricians usually do not pay enough attention to AAD.Domestic experts from pulmonary medicine, infection and gastroenterology are organized to develop the consensus, to improve the diagnosis, treatment and prevention of AAD, and contribute the children health in future.

Objective To compare the predictive value of CT severity index(CTSI)and modified CT severity index (MCTSI) for the severity and prognosis of acute pancreatitis (AP).Methods The clinical data of 72 patients with AP were retrospectively analyzed.The consistency between MCTSI,CTSI and revised Atlanta classification of AP-2012 was respectively compared by the Kappa test.The sensitivity,specificity,positive predictive value(PPV)and negative predictive value(NPV)of the MCTSI and CTSI in differentiating severe acute pancreatitis (SAP) were calculated.The predictive value of MCTSI and CTSI for SAP and mortality was respectively analyzed by receiver operating characteristic (ROC)curve,and the area under the curve(AUC)was respectively calculated.Results The MCTSI and CTSI showed good agreement with the revised Atlanta classification of AP-2012(the Kappa statistic was 0.653 and 0.656 respectively).The sensitivity,specificity,PPV and NPV of MCTSI were 90.0%(18/20), 86.5% (45/52), 72.0% (18/25) and 95.7% (45/47) respectively; and those of CTSI were 50.0% (10/20), 96.2% (50/52), 83.3% (10/12) and 83.3% (50/60) respectively.The AUC of MCTSI for predicting SAP and mortality were 0.932 and 0.913 respectively, and of CTSI were 0.908 and 0.874 respectively.Conclusions The MCTSI and CTSI have a similar and good consistency in predicting severity of AP.However,the MCTSI shows more sensitive and accurate in predicting SAP and prognosis.

Objective To evaluate the evidence and recommendations of the relevant clinical practice guidelines for the prevention and treatment of stroke published in the last five years. Methods The stroke and guideline-related key words and free words domestic computer retrieval platforms(China biomedical literature database,CNKI,and Yimaitong etc.),as well as foreign platforms(PubMed,NGC,GIN,etc.) were retrieved.The retrieval time limit ranged from January 2012 to May 2017.Appraisal of Guidelines for Research and Evaluation Ⅱ(AGREE Ⅱ)was used to assess the quality of the guidelines. Results A total of 9 490 documents were retrieved preliminarily and finally 16 articles were included. The average scoring rates of the 16 guidelines in the 6 fields of AGREE Ⅱ,including the scope and purpose,participants,rigor of the formulation,clarity of expression,applicability,and independence of the editors were 66.3%, 27.4%,23.0%,53.9%,18.8%,and 0%,respectively.The different guidelines all recommend that internal endarterectomy or carotid balloon dilatation and carotid artery stenting are used to extracranial carotid stenosis,while the treatment of vertebral artery stenting is recommended for extracranial vertebral artery stenosis.However,the intravascular interventional therapy is not recommended for intracranial arterial stenosis.The perioperative period of ischemic stroke intravascular intervention need to pay attention to the application of anti-platelet aggregation,anticoagulation,blood pressure,blood glucose,blood lipid management,and postoperative monitoring. For postoperative stent restenosis,especially after vertebral artery stenting,the guidelines recommend using drug-eluting stents,however,the dosage of heparin in intraoperative anticoagulation is still controversial.The intracranial aneurysm guidelines are more recommended for interventional therapy.The cerebral hemorrhage in different regions is recommended for different surgical procedures. Conclusions The guidelines in this field can reflect the key problems in clinical practice and keep up with the international concept.However,the quality of the methodology of the guidelines development needs to be improved. Some of them need to be updated and the recommendations should refer more evidence of the Chinese population.

Objective:To investigate the clinical effect of low molecular weight heparin sodium combined with antivenin in the treatment of severe and critical bite by Trimeresurus stejnegeri.Methods:The clinical data of 48 patients with severe or critical bite by Trimeresurus stejnegeri admitted to emergency department of Southeast Hospital Affiliated to Xiamen University from March 2017 to May 2019 were retrospectively analyzed. On the basis of early treatment of antivenom serum, internal administration and external application of Jidesheng snake tablet, and wound incision and detoxification, the patients were divided into heparin treatment group and non-heparin treatment group according to whether the low molecular heparin sodium was used or not. The patients in the two groups were compared in terms of gender, age, clinical classification, swelling degree of injured limbs, change of coagulation function index, bleeding of skin, mucous membrane or digestive tract, blood transfusion, local symptoms of bite, length of hospital stay and prognosis.Results:There was no significant difference in terms of gender, age, clinical classification or swelling degree of injured limbs between the two groups. On the 3rd day of treatment, the platelet count (PLT) in the heparin treatment group was significantly higher than that in the non-heparin treatment group [×10 9/L: 210.0 (160.0, 252.0) vs. 136.0 (104.0, 198.5), P < 0.05]. However, there was no significant difference in the four coagulation test results between the two groups. On the 6th day of treatment, the plasma thrombin time (TT) in the heparin treatment group was significantly shorter than that on the 3rd day of treatment [s: 30.3 (20.4, 37.0) vs. 34.7 (24.0, 73.4), P < 0.05], and the fibrinogen (FIB) in the heparin treatment group was significantly higher than that in the non-heparin treatment group [g/L: 0.60 (0.31, 1.07) vs. 0.20 (0.14, 0.60), P < 0.01]. The incidence of bleeding in the heparin treatment group was significantly lower than that in the non-heparin treatment group [21.7% (5/23) vs. 64.0% (16/25), P < 0.01]; 11 patients in the heparin treatment group and 18 patients in the non-heparin treatment group received blood transfusion and prothrombin complex supplement respectively. There was no significant difference in the length of hospital stay between the heparin group and non-heparin treatment group (days: 6.91±1.92 vs. 7.48±2.27, P > 0.05). The patients in both groups were followed up for 1 week to 1 month after treatment, and no death or local necrosis of skin and soft tissue was found. Conclusions:For the patients with severe and critical bite by Trimeresurus stejnegeri, on the basis of injection of antivenom serum, internal administration and external application of Jidesheng snake tablet, and wound incision and detoxification, early application of low molecular weight heparin sodium anticoagulation and other comprehensive treatment is helpful to improve limb swelling and inflammation, reduce blood transfusion, promote the recovery of coagulation function, and shorten the length of hospitalization.

Objective:To explore the effectiveness of a hysteroscopic endometrial lesion diagnosis model de-veloped based on deep learning(DL)algorithm combined with gradient-weighted class activation mapping(Grad-CAM)visualization technology.Methods:303 hysteroscopy videos(4781 images)of 291 patients who un-derwent hysteroscopy examination in the Department of Gynecology,Renmin Hospital of Wuhan University from June 1,2021 to December 31,2022 were selected.The dataset was divided into a training set(3703 images)and a test set(1078 images)by weight sampling method.After the training set was used for model learning and train-ing,two model architectures,residual neural network(ResNet18)and efficient neural network(EfficientNet-B0),were selected to verify the model in the test set by five-class and two-class classification tasks,respectively.Tak-ing histopathology as the gold standard,the diagnostic efficacy was evaluated to select the optimal model,and the Grad-CAM layer was embedded in the optimal model to output hysteroscopy images of Grad-CAM.Results:①In the five-class classification tasks,the accuracy of EfficientNet-B0 model(93.23％)was higher than that of Res-Net18 model(84.23％);the area under the curve(AUC)of EfficientNet-B0 model in the diagnosis of five disea-ses,including atypical endometrial hyperplasia,endometrial polyps,endometrial cancer,endometrial atypical hy-perplasia,and submucous myoma,was slightly higher than that of ResNet18 model,and the AUC of both models was almost above 0.980.②In the binary classification task of accuracy and the evaluation of specificity,the two models were similar,both above 93.00％,and the sensitivity of EfficientNet-B0 model(91.14％)was significantly better than that of ResNet18 model(77.22％).③EfficientNet-B0 model combined with Grad-CAM algorithm could identify the abnormal areas in the image.After biopsy and pathological examination,it was confirmed that about 95％of the marked areas in the model's output heatmap were lesion areas.Conclusions:The hysteroscopy di-agnostic model developed by EfficientNet-B0 model combined with Grad-CAM has high diagnostic accuracy,sen-sitivity,and specificity,and has application value in the diagnosis of endometrial lesions.

Objective:To explore the effectiveness of a hysteroscopic endometrial lesion diagnosis model de-veloped based on deep learning(DL)algorithm combined with gradient-weighted class activation mapping(Grad-CAM)visualization technology.Methods:303 hysteroscopy videos(4781 images)of 291 patients who un-derwent hysteroscopy examination in the Department of Gynecology,Renmin Hospital of Wuhan University from June 1,2021 to December 31,2022 were selected.The dataset was divided into a training set(3703 images)and a test set(1078 images)by weight sampling method.After the training set was used for model learning and train-ing,two model architectures,residual neural network(ResNet18)and efficient neural network(EfficientNet-B0),were selected to verify the model in the test set by five-class and two-class classification tasks,respectively.Tak-ing histopathology as the gold standard,the diagnostic efficacy was evaluated to select the optimal model,and the Grad-CAM layer was embedded in the optimal model to output hysteroscopy images of Grad-CAM.Results:①In the five-class classification tasks,the accuracy of EfficientNet-B0 model(93.23％)was higher than that of Res-Net18 model(84.23％);the area under the curve(AUC)of EfficientNet-B0 model in the diagnosis of five disea-ses,including atypical endometrial hyperplasia,endometrial polyps,endometrial cancer,endometrial atypical hy-perplasia,and submucous myoma,was slightly higher than that of ResNet18 model,and the AUC of both models was almost above 0.980.②In the binary classification task of accuracy and the evaluation of specificity,the two models were similar,both above 93.00％,and the sensitivity of EfficientNet-B0 model(91.14％)was significantly better than that of ResNet18 model(77.22％).③EfficientNet-B0 model combined with Grad-CAM algorithm could identify the abnormal areas in the image.After biopsy and pathological examination,it was confirmed that about 95％of the marked areas in the model's output heatmap were lesion areas.Conclusions:The hysteroscopy di-agnostic model developed by EfficientNet-B0 model combined with Grad-CAM has high diagnostic accuracy,sen-sitivity,and specificity,and has application value in the diagnosis of endometrial lesions.

Objective:To explore the effectiveness of a hysteroscopic endometrial lesion diagnosis model de-veloped based on deep learning(DL)algorithm combined with gradient-weighted class activation mapping(Grad-CAM)visualization technology.Methods:303 hysteroscopy videos(4781 images)of 291 patients who un-derwent hysteroscopy examination in the Department of Gynecology,Renmin Hospital of Wuhan University from June 1,2021 to December 31,2022 were selected.The dataset was divided into a training set(3703 images)and a test set(1078 images)by weight sampling method.After the training set was used for model learning and train-ing,two model architectures,residual neural network(ResNet18)and efficient neural network(EfficientNet-B0),were selected to verify the model in the test set by five-class and two-class classification tasks,respectively.Tak-ing histopathology as the gold standard,the diagnostic efficacy was evaluated to select the optimal model,and the Grad-CAM layer was embedded in the optimal model to output hysteroscopy images of Grad-CAM.Results:①In the five-class classification tasks,the accuracy of EfficientNet-B0 model(93.23％)was higher than that of Res-Net18 model(84.23％);the area under the curve(AUC)of EfficientNet-B0 model in the diagnosis of five disea-ses,including atypical endometrial hyperplasia,endometrial polyps,endometrial cancer,endometrial atypical hy-perplasia,and submucous myoma,was slightly higher than that of ResNet18 model,and the AUC of both models was almost above 0.980.②In the binary classification task of accuracy and the evaluation of specificity,the two models were similar,both above 93.00％,and the sensitivity of EfficientNet-B0 model(91.14％)was significantly better than that of ResNet18 model(77.22％).③EfficientNet-B0 model combined with Grad-CAM algorithm could identify the abnormal areas in the image.After biopsy and pathological examination,it was confirmed that about 95％of the marked areas in the model's output heatmap were lesion areas.Conclusions:The hysteroscopy di-agnostic model developed by EfficientNet-B0 model combined with Grad-CAM has high diagnostic accuracy,sen-sitivity,and specificity,and has application value in the diagnosis of endometrial lesions.

Objective:To explore the effectiveness of a hysteroscopic endometrial lesion diagnosis model de-veloped based on deep learning(DL)algorithm combined with gradient-weighted class activation mapping(Grad-CAM)visualization technology.Methods:303 hysteroscopy videos(4781 images)of 291 patients who un-derwent hysteroscopy examination in the Department of Gynecology,Renmin Hospital of Wuhan University from June 1,2021 to December 31,2022 were selected.The dataset was divided into a training set(3703 images)and a test set(1078 images)by weight sampling method.After the training set was used for model learning and train-ing,two model architectures,residual neural network(ResNet18)and efficient neural network(EfficientNet-B0),were selected to verify the model in the test set by five-class and two-class classification tasks,respectively.Tak-ing histopathology as the gold standard,the diagnostic efficacy was evaluated to select the optimal model,and the Grad-CAM layer was embedded in the optimal model to output hysteroscopy images of Grad-CAM.Results:①In the five-class classification tasks,the accuracy of EfficientNet-B0 model(93.23％)was higher than that of Res-Net18 model(84.23％);the area under the curve(AUC)of EfficientNet-B0 model in the diagnosis of five disea-ses,including atypical endometrial hyperplasia,endometrial polyps,endometrial cancer,endometrial atypical hy-perplasia,and submucous myoma,was slightly higher than that of ResNet18 model,and the AUC of both models was almost above 0.980.②In the binary classification task of accuracy and the evaluation of specificity,the two models were similar,both above 93.00％,and the sensitivity of EfficientNet-B0 model(91.14％)was significantly better than that of ResNet18 model(77.22％).③EfficientNet-B0 model combined with Grad-CAM algorithm could identify the abnormal areas in the image.After biopsy and pathological examination,it was confirmed that about 95％of the marked areas in the model's output heatmap were lesion areas.Conclusions:The hysteroscopy di-agnostic model developed by EfficientNet-B0 model combined with Grad-CAM has high diagnostic accuracy,sen-sitivity,and specificity,and has application value in the diagnosis of endometrial lesions.