Objective To explore the impact of preoperative fibrinogen levels on the 1-year adverse outcomes after endovascular revascularization in patients with diabetes complicated with lower extremity arteriosclerosis obliterans(LEASO).Methods We collected the baseline clinical data of 289 patients with diabetes complicated with LEASO,who were admitted to The First Affiliated Hospital of Xi'an Jiaotong University from May 2020 to December 2022 for endovascular revascularization.All patients were followed up for 13 to 24 months after interventional therapy,with the follow-up information including major adverse cardiovascular events(MACEs)such as all-cause death,acute myocardial infarction and acute stroke,as well as major adverse lower extremity events(MALEs)such as rest pain in the lower extremities,ulcers or skin defects,gangrene,reocclusion and amputation.A multivariable Cox regression model was used to analyze the related risk factors for adverse outcomes 1 year after endovascular revascularization in patients with diabetes complicated with LEASO,and receiver operating characteristic(ROC)curves were constructed to evaluate the predictive efficacy and optimal cutoff value of fibrinogen levels for endpoint events,and Kaplan-Meier survival curves were drawn.Sensitivity analysis was made to assess the differences in the impact of fibrinogen on endpoint events across various subgroups.Results We recruited a total of 289 patients(55 patients in MACEs and 234 in non-MACEs;68 patients in MALEs and 221 in non-MALEs),with a mean age of 67.6±9.3 years,including 215 males.Multivariate Cox regression analysis showed that elevated plasma fibrinogen was an independent risk factor for MACEs(HR=1.250,95％CI:1.053-1.484,P=0.011)and all-cause death(HR=1.297,95％CI:1.030-1.633,P=0.027)in the cohort followed up 1 year after interventional therapy,but had no significant impact on the occurrence of MALEs(P=0.625).Baseline plasma fibrinogen level 4.32 g/L was the optimal cutoff value for predicting MACEs(sensitivity=0.673,95％CI:0.582-0.767;specificity=0.688,95％CI:0.562-0.775)and all-cause death(sensitivity=0.679,95％CI:0.483-0.880;specificity=0.651,95％CI:0.465-0.755).The AUC for predicting MACEs and all-cause death after interventional therapy was 0.652(95％CI:0.564 2-0.739 1)and 0.619(95％CI:0.507-0.733),respectively.After a median follow-up of 14.03 months,patients with preoperative fibrinogen level ≥ 4.32 g/L had a significantly higher risk of MACEs and all-cause death compared to patients with preoperative fibrinogen＜4.32 g/L(P＜0.001),and there were no significant differences in different subgroups,including gender(male/female,interaction P=0.836),age(＜65 years/≥65 years,interaction P=0.211),smoking status(never smoked/current or former smoker,interaction P=0.779),chronic kidney disease(yes/no,interaction P=0.360),and heart failure(yes/no,interaction P=0.114).Conclusion Preoperative plasma fibrinogen≥4.32 g/L is an effective indicator for predicting MACEs and all-cause mortality following endovascular revascularization in patients with diabetes and LEASO.

Objective To analyze the correlation between lipid metabolism indexes,serum gamma-glutamyl transpeptidyase(γ-GGT)and coronary heart disease(CHD)complicated with coronary artery calcification(CAC).Methods A total of 300 CHD patients admitted in this study were divided into the CAC group(n=193)and the non-CAC group(n=107).Clinical data of the two groups were compared,including high density lipoprotein cholesterol(HDL-C),low density lipoprotein cholesterol(LDL-C),total cholesterol(TC),triglyceride(TG),Apolipoprotein A1(Apo-A1),Apo-B(APO-B)and γ-GGT.The influencing factors of CAC were analyzed by multiple Logistic factors.And a nomogram prediction model was established.Results The basic data of the two groups were compared.Patients of the CAC group was older,had higher proportion of patients with hypertension and diabetes,had higher levels of LDL-C,TC,Apo-B and γ-GGT and lower level of Apo-A1 than those of the non-CAC group(P＜0.05).The results of Logistic multivariate regression analysis showed that advanced age,combined history of diabetes,elevated LDL-C,TC,Apo-B and γ-GGT were risk factors of CHD complicated with CAC,while elevated Apo-A1 was the protective factor of CHD complicated with CAC(P＜0.05).The AUC of the constructed nomogram model was 0.880(95%CI:0.840-0.919),which showed good distinguishing ability.Conclusion CHD complicated with CAC is related to lipid metabolism and γ-GGT level.The nomogram model constructed based on influencing factors can be used for clinical early warning of CAC risk.

Objective To explore the impact of preoperative fibrinogen levels on the 1-year adverse outcomes after endovascular revascularization in patients with diabetes complicated with lower extremity arteriosclerosis obliterans(LEASO).Methods We collected the baseline clinical data of 289 patients with diabetes complicated with LEASO,who were admitted to The First Affiliated Hospital of Xi'an Jiaotong University from May 2020 to December 2022 for endovascular revascularization.All patients were followed up for 13 to 24 months after interventional therapy,with the follow-up information including major adverse cardiovascular events(MACEs)such as all-cause death,acute myocardial infarction and acute stroke,as well as major adverse lower extremity events(MALEs)such as rest pain in the lower extremities,ulcers or skin defects,gangrene,reocclusion and amputation.A multivariable Cox regression model was used to analyze the related risk factors for adverse outcomes 1 year after endovascular revascularization in patients with diabetes complicated with LEASO,and receiver operating characteristic(ROC)curves were constructed to evaluate the predictive efficacy and optimal cutoff value of fibrinogen levels for endpoint events,and Kaplan-Meier survival curves were drawn.Sensitivity analysis was made to assess the differences in the impact of fibrinogen on endpoint events across various subgroups.Results We recruited a total of 289 patients(55 patients in MACEs and 234 in non-MACEs;68 patients in MALEs and 221 in non-MALEs),with a mean age of 67.6±9.3 years,including 215 males.Multivariate Cox regression analysis showed that elevated plasma fibrinogen was an independent risk factor for MACEs(HR=1.250,95％CI:1.053-1.484,P=0.011)and all-cause death(HR=1.297,95％CI:1.030-1.633,P=0.027)in the cohort followed up 1 year after interventional therapy,but had no significant impact on the occurrence of MALEs(P=0.625).Baseline plasma fibrinogen level 4.32 g/L was the optimal cutoff value for predicting MACEs(sensitivity=0.673,95％CI:0.582-0.767;specificity=0.688,95％CI:0.562-0.775)and all-cause death(sensitivity=0.679,95％CI:0.483-0.880;specificity=0.651,95％CI:0.465-0.755).The AUC for predicting MACEs and all-cause death after interventional therapy was 0.652(95％CI:0.564 2-0.739 1)and 0.619(95％CI:0.507-0.733),respectively.After a median follow-up of 14.03 months,patients with preoperative fibrinogen level ≥ 4.32 g/L had a significantly higher risk of MACEs and all-cause death compared to patients with preoperative fibrinogen＜4.32 g/L(P＜0.001),and there were no significant differences in different subgroups,including gender(male/female,interaction P=0.836),age(＜65 years/≥65 years,interaction P=0.211),smoking status(never smoked/current or former smoker,interaction P=0.779),chronic kidney disease(yes/no,interaction P=0.360),and heart failure(yes/no,interaction P=0.114).Conclusion Preoperative plasma fibrinogen≥4.32 g/L is an effective indicator for predicting MACEs and all-cause mortality following endovascular revascularization in patients with diabetes and LEASO.

Objective To analyze the correlation between lipid metabolism indexes,serum gamma-glutamyl transpeptidyase(γ-GGT)and coronary heart disease(CHD)complicated with coronary artery calcification(CAC).Methods A total of 300 CHD patients admitted in this study were divided into the CAC group(n=193)and the non-CAC group(n=107).Clinical data of the two groups were compared,including high density lipoprotein cholesterol(HDL-C),low density lipoprotein cholesterol(LDL-C),total cholesterol(TC),triglyceride(TG),Apolipoprotein A1(Apo-A1),Apo-B(APO-B)and γ-GGT.The influencing factors of CAC were analyzed by multiple Logistic factors.And a nomogram prediction model was established.Results The basic data of the two groups were compared.Patients of the CAC group was older,had higher proportion of patients with hypertension and diabetes,had higher levels of LDL-C,TC,Apo-B and γ-GGT and lower level of Apo-A1 than those of the non-CAC group(P＜0.05).The results of Logistic multivariate regression analysis showed that advanced age,combined history of diabetes,elevated LDL-C,TC,Apo-B and γ-GGT were risk factors of CHD complicated with CAC,while elevated Apo-A1 was the protective factor of CHD complicated with CAC(P＜0.05).The AUC of the constructed nomogram model was 0.880(95%CI:0.840-0.919),which showed good distinguishing ability.Conclusion CHD complicated with CAC is related to lipid metabolism and γ-GGT level.The nomogram model constructed based on influencing factors can be used for clinical early warning of CAC risk.

Objective To explore the effect of different treatment methods on prognosis in elderly patients with lower extremity arterial occlusive disease.Methods A total of 352 elderly patients with lower extremity arterial occlusive disease admitted in our hospital from May 2020 to May 2022 were enrolled,and according to their willingness and characteristics of lower extremity le-sions,they were divided into balloon dilation group(142 patients),stent implantation group(145 patients)and conservative treatment group(65 patients).All patients were followed up for 13-24 months.The incidences of major adverse cardiovascular events(MACE),including all-cause death,acute myocardial infarction,acute ischemic stroke,and major adverse lower limb events(MALE),including lower extremity pain at rest,ulcers or skin defects,gangrene,reocclusion,and amputation were observed and recorded.The clinical data and prognosis were compared and ana-lyzed of the three groups.Kaplan-Meier survival curves were drawn.Results The incidence of all-cause mortality was significantly lower in the stent implantation group than the conservative treatment group(9.7％vs 23.1％,P＜0.01).The incidence of MALE was obviously lower in the stent implantation group and the balloon dilatation group than the conservative treatment group(4.8％and 9.2％vs 24.6％,P＜0.01).Conclusion Endovascular therapy can reduce the risk of all-cause death and MALE occurrence in elderly patients with lower extremity arterial occlusive disease who are suitable for interventional therapy.

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.

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.