Objective:To evaluate the value of the proposed interpretable denoising deep learning model-grouped sharing convolutional attention-visual transformer（GSCAViT）for classifying normal fetal echocardiograms.Methods:A total of 2 501 images from 358 fetuses who underwent cardiac ultrasound examinations at Xiangya Second Hospital of Central South University from January to November 2024 were retrospectively analyzed. GSCAViT was constructed based on fetal echocardiograms from the three-vessel and trachea view，apical four-chamber view，long-axis view of the aortic arch，bicaval view，left ventricular outflow tract view，three-vessel view and right ventricular outflow tract view were compared with both baseline and improved models in the validation set to assess the performance of the classification echocardiography in terms of accuracy，precision，recall and F1-score. Its generalizability across test sets was assessed using the area under the ROC curve（AUC），sensitivity，specificity and F1-score. The impact of image features was interpreted using SHapley Additive exPlanations（SHAP）.The effectiveness of the GSCA module was compared through visual analysis，image parameter metrics and classification performance.Results:The GSCAViT model achieved classification performance for fetal echocardiograms second only to MaxViT，with an accuracy of 97.1%，precision of 97.1%，recall of 97.0%，and an F1-score of 97.0%. In the E10，E20 and E8 test sets，the AUCs of GSCAViT for the prediction of 7 types of fetal echocardiograms were 0.994，0.928 and 0.932，the sensitivities were 99.4%，81.3% and 72.9%，the specificities were 99.7%，96.8% and 94.8%，the F1-scores were 99.4%，81.3% and 67.6%，respectively. SHAP visualization indicated that the model could identify key structural features within the images. Images processed by the denoising-guided group-sharing convolutional attention module best captured and enhanced important regional features，achieving the highest contrast-to-noise ratio，peak signal-to-noise ratio and optimal classification performance，which demonstrated the module's effectiveness.Conclusions:The proposed GSCAViT model exhibits superior performance in classifying seven types of normal fetal echocardiograms compared to the baseline and some improved models. Furthermore，SHAP visualization enhances the interpretability of the classification results，and visual comparisons，image parameter analyses，as well as classification performance metrics confirming the effectiveness of the denoising-guided group-sharing convolutional attention module in the visual transformer model.

Objective:To evaluate the value of the proposed interpretable denoising deep learning model-grouped sharing convolutional attention-visual transformer（GSCAViT）for classifying normal fetal echocardiograms.Methods:A total of 2 501 images from 358 fetuses who underwent cardiac ultrasound examinations at Xiangya Second Hospital of Central South University from January to November 2024 were retrospectively analyzed. GSCAViT was constructed based on fetal echocardiograms from the three-vessel and trachea view，apical four-chamber view，long-axis view of the aortic arch，bicaval view，left ventricular outflow tract view，three-vessel view and right ventricular outflow tract view were compared with both baseline and improved models in the validation set to assess the performance of the classification echocardiography in terms of accuracy，precision，recall and F1-score. Its generalizability across test sets was assessed using the area under the ROC curve（AUC），sensitivity，specificity and F1-score. The impact of image features was interpreted using SHapley Additive exPlanations（SHAP）.The effectiveness of the GSCA module was compared through visual analysis，image parameter metrics and classification performance.Results:The GSCAViT model achieved classification performance for fetal echocardiograms second only to MaxViT，with an accuracy of 97.1%，precision of 97.1%，recall of 97.0%，and an F1-score of 97.0%. In the E10，E20 and E8 test sets，the AUCs of GSCAViT for the prediction of 7 types of fetal echocardiograms were 0.994，0.928 and 0.932，the sensitivities were 99.4%，81.3% and 72.9%，the specificities were 99.7%，96.8% and 94.8%，the F1-scores were 99.4%，81.3% and 67.6%，respectively. SHAP visualization indicated that the model could identify key structural features within the images. Images processed by the denoising-guided group-sharing convolutional attention module best captured and enhanced important regional features，achieving the highest contrast-to-noise ratio，peak signal-to-noise ratio and optimal classification performance，which demonstrated the module's effectiveness.Conclusions:The proposed GSCAViT model exhibits superior performance in classifying seven types of normal fetal echocardiograms compared to the baseline and some improved models. Furthermore，SHAP visualization enhances the interpretability of the classification results，and visual comparisons，image parameter analyses，as well as classification performance metrics confirming the effectiveness of the denoising-guided group-sharing convolutional attention module in the visual transformer model.

Objective:To establish the reference value of the angle between the left ventricular inflow and outflow tract (LIOA) in normal fetuses in the second and third trimesters, and observe the correlation between fetal LIOA and gestational age, cardiac axis, cardiac size, and estimated fetal weigh (EFW).Methods:Fetal LIOA in normal fetuses with gestational age from 16 weeks to 39 + 6 weeks were obtained prospectively by two-dimensional ultrasound in the Second Xiangya Hospital from November 2022 to April 2023. Pearson′s correlation coefficient was used to analyze the correlation between fetal LIOA and gestational age, cardiac axis, cardiovascural size and EFW. Results:The LIOA range of 651 normal fetuses was (44.39±7.67)°, and it was found that LIOA was not related to gestational age. LIOA mildly positively correlated with the cardiac axis ( r=0.22, P<0.05) while not correlated with gestational age, cardiovascural size or EFW (all P>0.05). Conclusions:The range of LIOA in normal fetuses were established. Fetal LIOA is constant in the second and third trimesters and it is mildly positively correlated with the cardiac axis. Evaluating fetal LIOA may also provide information for future research on the fetal aortic hemodynamic development.

ObjectiveTo investigate the efficacy of rapamycin combined with CsA/Tacrolimus (Tac) in chronic allograft nephropathy (CAN).MethodsFifty-three cases of CAN accepted the quadruple immunosuppressive drug program,which contained rapamycin combined with CsA/Tac and MMF and prednisone,and CsA/Tac and MMF were reduced to the original amount of 25％ to 50％.After treatment for 12 months,more relevant indicators,including serum creatinine,glomerular filtration rate,serum cholesterol,triglycerides,urinary protein,GPT and bilirubin and other changes were observed.ResultsIn the patients receiving quadruple regimen of rapamycin during 12 months,the blood Ccr was decreased from (161.51 ± 106.48)μmol/L before treatment to (126.51 ± 56.2)μmol/L after treatment for 6 months (P＜0.05) and to (123.43 ± 54.18)μmol/L after for 12 months (P＜0.01).The GFR was increased from (0.754 ± 0.302) ml/s before treatment to (0.952 ± 0.347)ml/s after treatment for 6 months (P＜0.05) and to (1.007 ± 0.394) ml/s after treatment for 12 months (P＜0.01).Cholesterol and triglycerides in patients had no significant change before and after treatment.The positive rate of proteinuria after treatment showed an increasing trend from 9.4％ before treatment to 26.4％ after treatment for 12 months.ConclusionThe quadruple program of rapamycin combined with CsA/FK506 and MMF can significantly improve Ccr and GFR in patients with CAN,but it can increase the incidence of proteinuria in patients: