Objective To investigate the value of polar residual network(PResNet)model for assisting evaluation on rat myocardial infarction(MI)segment in myocardial contrast echocardiography(MCE).Methods Twenty-five male SD rats...Objective To investigate the value of polar residual network(PResNet)model for assisting evaluation on rat myocardial infarction(MI)segment in myocardial contrast echocardiography(MCE).Methods Twenty-five male SD rats were randomly divided into MI group(n=15)and sham operation group(n=10).MI models were established in MI group through ligation of the left anterior descending coronary artery using atraumatic suture,while no intervention was given to those in sham operation group after thoracotomy.MCE images of both basal and papillary muscle levels on the short axis section of left ventricles were acquired after 1 week,which were assessed independently by 2 junior and 2 senior ultrasound physicians.The evaluating efficacy of MI segment,the mean interpretation time and the consistency were compared whether under the assistance of PResNet model or not.Results No significant difference of efficacy of evaluation on MI segment was found for senior physicians with or without assistance of PResNet model(both P>0.05).Under the assistance of PResNet model,the efficacy of junior physicians for diagnosing MI segment was significantly improved compared with that without the assistance of PResNet model(both P<0.01),and was comparable to that of senior physicians.Under the assistance of PResNet model,the mean interpretation time of each physician was significantly shorter than that without assistance(all P<0.001),and the consistency between junior physicians and among junior and senior physicians were both moderate(Kappa=0.692,0.542),which became better under the assistance(Kappa=0.763,0.749).Conclusion PResNet could improve the efficacy of junior physicians for evaluation on rat MI segment in MCE images,shorten interpretation time with different aptitudes,also improve the consistency to some extent.展开更多
Objective To observe the value of deep learning (DL) models for automatic classification of echocardiographic views. Methods Totally 100 patients after heart transplantation were retrospectively enrolled and divided i...Objective To observe the value of deep learning (DL) models for automatic classification of echocardiographic views. Methods Totally 100 patients after heart transplantation were retrospectively enrolled and divided into training set, validation set and test set at a ratio of 7 ∶ 2 ∶ 1. ResNet18, ResNet34, Swin Transformer and Swin Transformer V2 models were established based on 2D apical two chamber view, 2D apical three chamber view, 2D apical four chamber view, 2D subcostal view, parasternal long-axis view of left ventricle, short-axis view of great arteries, short-axis view of apex of left ventricle, short-axis view of papillary muscle of left ventricle, short-axis view of mitral valve of left ventricle, also 3D and CDFI views of echocardiography. The accuracy, precision, recall, F1 score and confusion matrix were used to evaluate the performance of each model for automatically classifying echocardiographic views. The interactive interface was designed based on Qt Designer software and deployed on the desktop. Results The performance of models for automatically classifying echocardiographic views in test set were all good, with relatively poor performance for 2D short-axis view of left ventricle and superior performance for 3D and CDFI views. Swin Transformer V2 was the optimal model for automatically classifying echocardiographic views, with high accuracy, precision, recall and F1 score was 92.56%, 89.01%, 89.97% and 89.31%, respectively, which also had the highest diagonal value in confusion matrix and showed the best classification effect on various views in t-SNE figure. Conclusion DL model had good performance for automatically classifying echocardiographic views, especially Swin Transformer V2 model had the best performance. Using interactive classification interface could improve the interpretability of prediction results to some extent.展开更多
Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-cham...Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-chamber views two-dimensional echocardiograms were obtained prospectively in 205 patients with coronary heart disease.The model for evaluating LV regional contractile function was constructed using a five-fold cross-validation method to automatically identify the presence of RWMA or not,and the performance of this model was assessed taken manual interpretation of RWMA as standards.Results Among 205 patients,RWMA was detected in totally 650 segments in 83 cases.LV myocardial segmentation model demonstrated good efficacy for delineation of LV myocardium.The average Dice similarity coefficient for LV myocardial segmentation results in the apical two-chamber,three-chamber and four-chamber views was 0.85,0.82 and 0.88,respectively.LV myocardial segmentation model accurately segmented LV myocardium in apical two-chamber,three-chamber and four-chamber views.The mean area under the curve(AUC)of RWMA identification model was 0.843±0.071,with sensitivity of(64.19±14.85)%,specificity of(89.44±7.31)%and accuracy of(85.22±4.37)%.Conclusion Deep learning echocardiographic intelligent model could be used to automatically evaluate LV regional contractile function,hence rapidly and accurately identifying RWMA.展开更多
A new approach was taken to investigate the crustal stucture of the Kane transform and its aseismic extension, using high quality bathymetry and gravity data. The gravity signatures associated with variations in crust...A new approach was taken to investigate the crustal stucture of the Kane transform and its aseismic extension, using high quality bathymetry and gravity data. The gravity signatures associated with variations in crustal thickness of the transform were isolated from the observed free air anomaly, was continued downward to the mean depth of the crust/mantle interface and converted onto the relief on that surface. The crustal thickness of the transform was then calculated by subtracting seawater depth from the depth of the gravity inferred crust/mantle interface.3 D gravity investigation results indicate that the Kane transform and adjacent areas are associated with a crust thinner than normal oceanic crust. The transform trough is largely underlain by a crust less than 4.5km thick and in the nodal basins the crust may be as thin as 3 km. The crust beneath the fracture zone valley is 4-5.5 km thick. The rift valleys on the spreading segments are also characterized by thin crust (4-5 km thick). Thin oceanic crust extends to 20-30 km from the transform axis,except for some localized places such as the inside corner highs adjoining the ridge transform intersections. These gravity inferred results match fairly well with limited published seismic results. Thinning of the crust is mainly attributable to a thin layer 3, which in turn may be explained by the combined effects of reduced magma supply at the ends of the spreading segments and tectonic activities in the region.展开更多
The possible collapse of different circumstances is derived with the help of the limit analysis theory.Analytical equations related to collapsing mechanisms in deep tunnel with smooth three-centered arc cross sections...The possible collapse of different circumstances is derived with the help of the limit analysis theory.Analytical equations related to collapsing mechanisms in deep tunnel with smooth three-centered arc cross sections are derived on the basis of Hoek-Brown failure criterion and upper bound limit analysis.The pore water pressure is considered in the analysis,as a work rate of external force.Numerical results about the shape of detaching curve and the weight of collapsing block per unit length corresponding to different parameters are obtained with the help of mathematical software.The shapes of collapsing block are drawn with respected to different parameters.Furthermore,the effects of different parameters on the shape of detaching curve and the weight of the collapsing block are discussed.展开更多
[ Objective] The aim was to discuss the optimum fermentation technology of kiwi wine. [Method] Effects of pH, SO2 concentration, fer- mentation temperature, yeast inoculums amount on the quality of kiwi wine were inve...[ Objective] The aim was to discuss the optimum fermentation technology of kiwi wine. [Method] Effects of pH, SO2 concentration, fer- mentation temperature, yeast inoculums amount on the quality of kiwi wine were investigated by single factor experiment and orthogonal experiment. P, esult] The optimum conditions were as follows： fermentation temperature （22 ℃ ）, yeast inoculums （0.20 g/L）, SO2 concentration （60 mg/L） and pH （3.5）. The kiwi wine produced under these conditions was yellowish green, transparent, moderately acidic with strong fruit flavor and aro- ma. [ Conclusion]The study provided a theoretical foundation for the development and production of deeply processed products of kiwifruit.展开更多
文摘Objective To investigate the value of polar residual network(PResNet)model for assisting evaluation on rat myocardial infarction(MI)segment in myocardial contrast echocardiography(MCE).Methods Twenty-five male SD rats were randomly divided into MI group(n=15)and sham operation group(n=10).MI models were established in MI group through ligation of the left anterior descending coronary artery using atraumatic suture,while no intervention was given to those in sham operation group after thoracotomy.MCE images of both basal and papillary muscle levels on the short axis section of left ventricles were acquired after 1 week,which were assessed independently by 2 junior and 2 senior ultrasound physicians.The evaluating efficacy of MI segment,the mean interpretation time and the consistency were compared whether under the assistance of PResNet model or not.Results No significant difference of efficacy of evaluation on MI segment was found for senior physicians with or without assistance of PResNet model(both P>0.05).Under the assistance of PResNet model,the efficacy of junior physicians for diagnosing MI segment was significantly improved compared with that without the assistance of PResNet model(both P<0.01),and was comparable to that of senior physicians.Under the assistance of PResNet model,the mean interpretation time of each physician was significantly shorter than that without assistance(all P<0.001),and the consistency between junior physicians and among junior and senior physicians were both moderate(Kappa=0.692,0.542),which became better under the assistance(Kappa=0.763,0.749).Conclusion PResNet could improve the efficacy of junior physicians for evaluation on rat MI segment in MCE images,shorten interpretation time with different aptitudes,also improve the consistency to some extent.
文摘Objective To observe the value of deep learning (DL) models for automatic classification of echocardiographic views. Methods Totally 100 patients after heart transplantation were retrospectively enrolled and divided into training set, validation set and test set at a ratio of 7 ∶ 2 ∶ 1. ResNet18, ResNet34, Swin Transformer and Swin Transformer V2 models were established based on 2D apical two chamber view, 2D apical three chamber view, 2D apical four chamber view, 2D subcostal view, parasternal long-axis view of left ventricle, short-axis view of great arteries, short-axis view of apex of left ventricle, short-axis view of papillary muscle of left ventricle, short-axis view of mitral valve of left ventricle, also 3D and CDFI views of echocardiography. The accuracy, precision, recall, F1 score and confusion matrix were used to evaluate the performance of each model for automatically classifying echocardiographic views. The interactive interface was designed based on Qt Designer software and deployed on the desktop. Results The performance of models for automatically classifying echocardiographic views in test set were all good, with relatively poor performance for 2D short-axis view of left ventricle and superior performance for 3D and CDFI views. Swin Transformer V2 was the optimal model for automatically classifying echocardiographic views, with high accuracy, precision, recall and F1 score was 92.56%, 89.01%, 89.97% and 89.31%, respectively, which also had the highest diagonal value in confusion matrix and showed the best classification effect on various views in t-SNE figure. Conclusion DL model had good performance for automatically classifying echocardiographic views, especially Swin Transformer V2 model had the best performance. Using interactive classification interface could improve the interpretability of prediction results to some extent.
文摘Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-chamber views two-dimensional echocardiograms were obtained prospectively in 205 patients with coronary heart disease.The model for evaluating LV regional contractile function was constructed using a five-fold cross-validation method to automatically identify the presence of RWMA or not,and the performance of this model was assessed taken manual interpretation of RWMA as standards.Results Among 205 patients,RWMA was detected in totally 650 segments in 83 cases.LV myocardial segmentation model demonstrated good efficacy for delineation of LV myocardium.The average Dice similarity coefficient for LV myocardial segmentation results in the apical two-chamber,three-chamber and four-chamber views was 0.85,0.82 and 0.88,respectively.LV myocardial segmentation model accurately segmented LV myocardium in apical two-chamber,three-chamber and four-chamber views.The mean area under the curve(AUC)of RWMA identification model was 0.843±0.071,with sensitivity of(64.19±14.85)%,specificity of(89.44±7.31)%and accuracy of(85.22±4.37)%.Conclusion Deep learning echocardiographic intelligent model could be used to automatically evaluate LV regional contractile function,hence rapidly and accurately identifying RWMA.
文摘A new approach was taken to investigate the crustal stucture of the Kane transform and its aseismic extension, using high quality bathymetry and gravity data. The gravity signatures associated with variations in crustal thickness of the transform were isolated from the observed free air anomaly, was continued downward to the mean depth of the crust/mantle interface and converted onto the relief on that surface. The crustal thickness of the transform was then calculated by subtracting seawater depth from the depth of the gravity inferred crust/mantle interface.3 D gravity investigation results indicate that the Kane transform and adjacent areas are associated with a crust thinner than normal oceanic crust. The transform trough is largely underlain by a crust less than 4.5km thick and in the nodal basins the crust may be as thin as 3 km. The crust beneath the fracture zone valley is 4-5.5 km thick. The rift valleys on the spreading segments are also characterized by thin crust (4-5 km thick). Thin oceanic crust extends to 20-30 km from the transform axis,except for some localized places such as the inside corner highs adjoining the ridge transform intersections. These gravity inferred results match fairly well with limited published seismic results. Thinning of the crust is mainly attributable to a thin layer 3, which in turn may be explained by the combined effects of reduced magma supply at the ends of the spreading segments and tectonic activities in the region.
基金Project(2013CB036004)supported by Basic Research Program of ChinaProjects(51378510,41302226)supported by National Natural Science Foundation of ChinaProject(2016zzts062)supported by Doctorial Innovation Foundation of Central South University,China
文摘The possible collapse of different circumstances is derived with the help of the limit analysis theory.Analytical equations related to collapsing mechanisms in deep tunnel with smooth three-centered arc cross sections are derived on the basis of Hoek-Brown failure criterion and upper bound limit analysis.The pore water pressure is considered in the analysis,as a work rate of external force.Numerical results about the shape of detaching curve and the weight of collapsing block per unit length corresponding to different parameters are obtained with the help of mathematical software.The shapes of collapsing block are drawn with respected to different parameters.Furthermore,the effects of different parameters on the shape of detaching curve and the weight of the collapsing block are discussed.
文摘[ Objective] The aim was to discuss the optimum fermentation technology of kiwi wine. [Method] Effects of pH, SO2 concentration, fer- mentation temperature, yeast inoculums amount on the quality of kiwi wine were investigated by single factor experiment and orthogonal experiment. P, esult] The optimum conditions were as follows： fermentation temperature （22 ℃ ）, yeast inoculums （0.20 g/L）, SO2 concentration （60 mg/L） and pH （3.5）. The kiwi wine produced under these conditions was yellowish green, transparent, moderately acidic with strong fruit flavor and aro- ma. [ Conclusion]The study provided a theoretical foundation for the development and production of deeply processed products of kiwifruit.
