Objective:To investigate the clinical value of abdominal adipose volume in predicting early tumor recurrence after resection of hepatocellular carcinoma (HCC).Methods:The retrospective case-control study was conducted. The clinicopathological data of 132 HCC patients with tumor diameter ≤5 cm who were admitted to The First Affiliated Hospital of Army Medical University from December 2017 to October 2019 were collected. There were 110 males and 22 females, aged (51±4)years. All patients underwent resection of HCC. Preoperative computer tomography scanning was performed and the visceral and subcutaneous fats of patients were quantified using the Mimics Research 21.0 software. Based on time to postoperative tumor recurrence patients were divided to two categories: early recurrence and non-early recurrence. Observation indicators: (1) consistency analy-sis; (2) analysis of factors influencing early tumor recurrence after resection of HCC and construction of prediction model. Measurement data with normal distribution were represented as Mean± SD, and comparison between groups was conducted using the t test. Measurement data with skewed distribu-tion were represented as M( Q1,Q3) or M(range), and comparison between groups was conducted using the Mann-Whitney U test. Count data were expressed as absolute numbers, and comparison between groups was conducted using the chi-square test or Fisher exact probability. Consistency analysis was conducted using the intragroup correlation coefficient (ICC) test. Multivariate analysis was performed using the binary Logistic regression model forward method. Independent risk factors influencing early tumor recurrence after resection of HCC were screened. The area under curve (AUC) of receiver operating characteristic (ROC) curve was applied to select the optimal cut-off value to classify high and low risks of recurrence. The Kaplan-Meier method was used to draw survival curve and calculate survival time. The Log-Rank test was used for survival analysis. Results:(1) Consistency analysis. The consistency ICC of abdominal fat parameters of visceral fat volume (VFV), subcutaneous fat volume, visceral fat area, and subcutaneous fat area measured by 2 radiologists were 0.84, 1.00, 0.86, and 0.94, respectively. (2) Analysis of factors influencing early tumor recurr-ence after resection of HCC and construction of prediction model. All 132 patients were followed up after surgery for 662(range, 292-1 111)days. During the follow-up, there were 52 patients with non-early recurrence and 80 patients with early recurrence. Results of multivariate analysis showed that VFV was an independent factor influencing early tumor recurrence after resection of HCC ( odds ratio=4.07, 95% confidence interval as 2.27-7.27, P<0.05). The AUC of ROC curve based on VFV was 0.78 (95% confidence interval as 0.70-0.85), and the sensitivity and specificity were 72.2 % and 77.4 %, respectively. The optimal cut-off value of VFV was 1.255 dm 3, and all 132 patients were divided into the high-risk early postoperative recurrence group of 69 cases with VFV >1.255 dm 3, and the low-risk early postoperative recurrence group of 63 cases with VFV ≤1.255 dm 3. The disease-free survival time of the high-risk early postoperative recurrence group and the low-risk early post-operative recurrence group were 414(193,702)days and 1 047(620,1 219)days, showing a significant difference between them ( χ2=31.17, P<0.05). Conclusions:VFV is an independent factor influen-cing early tumor recurrence of HCC after resection. As a quantitative indicator of abdominal fat, it can predict the prognosis of HCC patients.

Objective:To explore the prognostic value of baseline 18F-FDG PET/CT metabolic parameters in locally advanced cervical cancer (LACC) after concurrent chemoradiotherapy (CCRT). Methods:From September 2015 to October 2021, the clinical data of 180 LACC patients (age: 22-76 years) who underwent 18F-FDG PET/CT before CCRT at Affiliated Cancer Hospital of Shandong First Medical University were analyzed retrospectively. The metabolic tumor volume (MTV), total lesion glycolysis (TLG), SUV max, and SUV mean were computed by using the margin threshold of 42%SUV max. The optimal threshold for predicting progression-free survival (PFS) was obtained by ROC curve analysis. The Kaplan-Meier method was applied for survival analysis, and the log-rank test was applied to compare the survival rate between groups. Multivariate Cox proportional hazard regression was used to analyze progression for PFS. Results:The median follow-up was 19.1 months, and 54 patients (30.0%, 54/180) suffered from disease progression. ROC curve analysis showed that the optimal cut-off value of MTV was 31.145 ml, with the AUC of 0.641. Para-aortic lymph node (PALN) metastasis had the highest AUC value (0.589) among the clinical factors, followed by International Federation of Gynecology and Obstetrics (FIGO) stage (0.581). The 1-year PFS rates of patients with MTV<31.145 ml ( n=88) and MTV≥31.145 ml ( n=92) were 80.68% and 59.78%, respectively ( χ2=13.72, P<0.001). Multivariate Cox analysis demonstrated that pathological type (hazard ratio ( HR)=3.075, 95% CI: 1.370-6.901, P=0.006), FIGO stage ( HR=1.955, 95% CI: 1.031-3.707, P=0.040), PALN metastasis ( HR=2.136, 95% CI: 1.202-3.796, P=0.010) and MTV ( HR=2.449, 95% CI: 1.341-4.471, P=0.004) were the significant predictors for PFS. Conclusions:Pathological type, FIGO stage, PALN metastasis and MTV are independent prognostic risk factors for PFS. MTV as the baseline 18F-FDG PET/CT metabolic parameter, can realize prognostic stratification analysis.

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.

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.

ObjectiveTo investigate the possible mechanism of electroacupuncture at "Neiguan" （PC6） and "Zhongwan" （RN12） in the treatment of functional dyspepsia （FD）. MethodsSPF male SD rats were randomly divided into model group， Neiguan （PC6） group， Zhongwan （RN12） group， and Neiguan-Zhongwan （PC6-RN12） group， with 8 rats in each group. Except for the normal group， the other groups of rats were cogavaged with 0.1% sucrose iodoacetamide solution combined with small platform standing training to establish FD rat models. After successful modeling， the rats in the normal group and model group were tied up for 30 min/d for 7 days； the Neiguan group， Zhongwan group， and Neiguan-Zhongwan group were treated with electroacupuncture intervention at "Neiguan" （PC6）， "Zhongwan" （RN12）， and "Neiguan" - "Zhongwan" （PC6-RN12） acupoints， respectively， using continuous wave， frequency of 2 Hz， current intensity of 1 mA， 30 min/d， and continuous intervention for 7 days. The general condition of rats in each group was observed. After treatment， the body weight and food intake of rats were measured， and the gastric emptying rate was calculated； HE staining was performed on the gastric antrum tissue of rats to observe the histopathologic changes； the expressions of calcitonin gene-related peptide （CGRP） and Ghrelin protein in gastric antrum were detected by Western Blot； the metabolites in gastric antrum tissues were detected by liquid chromatography-mass spectrometry （LC-MS）， and differential metabolites were screened by correlation analysis， then Metabo Analyst 5.0 and KEGG databases were used for metabolic pathway analysis. ResultsUnder light microscope， the gastric antrum structure was complete and the glands were abundant. No obvious inflammation and edema were found in gastric mucosa. Compared with the normal group， the body weight， food intake， and gastric emptie rate of rats in model group decreased， the expression of Ghrelin protein decreased and the expression of CGRP protein increased in gastric antrum tissue （P＜0.01）. Compared with model group， the body weight， food intake， and gastric emptance rate of rats in Neiguan group， Zhongwan group and Neiguan-Zhongwan group all increased， CGRP protein expression decreased in Neiguan group， and Ghrelin protein expression increased and CGRP protein expression decreased in Zhongwan group and Neiguan-Zhongwan group （P＜0.01 or P＜0.05）. Compared with Neiguan-Zhongwan group， Ghrelin protein expression decreased and CGRP protein expression increased in Neiguan group and Zhongwan group （P＜0.05 or P＜0.01）. Metabolomics results showed that compared with normal group， the content of metabolites adenosine diphosphate ribose， adenosine monophosphate and adenosine diphosphate in gastric antrum tissue of model group decreased； compared with model group， the contents of adenosine diphosphate， adenosine diphosphate and citicoline in Neiguan group increased， the contents of nicotine adenine dinucleotide， cytidine diphosphate ethanolamine and citicoline in Zhongwan group increased， and the contents of adenosine diphosphate， cytidine diphosphate and citicoline in Neiguan-Zhongwan group increased （P＜0.05 or P＜0.01）. Compared with model group， the main metabolic pathways of different metabolites in PC6-RN12 group were glycerophospholipid metabolism， taurine and hypotaurine metabolism， purine metabolism， starch and sucrose metabolism. ConclusionElectroacupuncture at “Neiguan” and “Zhongwan” acupoints can effectively regulate gastrointestinal motility and improve FD symptoms in FD rats， and the effect is better than that of "Neiguan" or "Zhongwan" acupoints alone. The mechanism may be related to the influence of related metabolites on energy metabolism， glucose metabolism and nucleotide metabolism， thereby regulating gastrointestinal motility hormones.