To explore the correlation between the left atrial appendage orifice diameter parameters measured by three-dimensional transesophageal echocardiography (3D-TEE) and the size of the Watchman left atrial appendage occluder in patients with atrial fibrillation.

Seventy-five patients with non-valvular atrial fibrillation who were scheduled to undergo Watchman left atrial appendage occlusion at the Department of Cardiovascular Medicine of the Second Hospital of Hebei Medical University from June 2020 to June 2022 were selected, and TEE examinations were completed within 48 hours before the operation. The inner diameter of the left atrial appendage at 0°, 45°, 90°, and 135° sections was measured by two-dimensional (2D) TEE. The maximum and minimum values of the left atrial appendage opening, the left atrial appendage perimeter derived diameter (PDD), and the left atrial appendage area derived diameter (ADD) were measured by 3D-TEE. Spearman correlation and linear regression analyses were performed between the above parameters and the model of left atrial appendage occluder implanted during the operation and the average diameter of the occluder after occluding.

All the patients were successfully implanted with a Watchman left atrial appendage occlusion device and completed TEE follow-up at 3 and 6 months after surgery. In the preoperative TEE examination, the inner diameter of the left atrial appendage opening in the group with reduced left atrial appendage emptying velocity (<40 cm/s) was larger than that of the group with normal left atrial appendage emptying velocity (≥40 cm/s). In the measurement of each section of 2D-TEE, the opening of left atrial appendage measured at 135° section was the largest, while the maximum opening of left atrial appendage measured by 3D-TEE was greater than that of 2D-TEE. During intraoperative and postoperative TEE follow-up, 17 cases (17/75, 22.60%) had complications. Spearman correlation analysis showed that the inner diameter of the left atrial appendage measured by 2D-TEE at 0°, 45°, 90°, and 135° sections as well as the maximum and minimum values of of the left atrial appendage opening, ADD, and PDD measured by 3D-TEE were positively correlated with the average diameter of the device after release, among which PDD had the highest correlation with the average diameter of the device after release (r_s=0.614, 0.660, 0.660, 0.704, 0.770, 0.624, 0.812, and 0.825, respectively, P<0.001). The inner diameter of the left atrial appendage measured by 2D-TEE at 0°, 45°, 90°, and 135° sections as well as the maximum and minimum values of of the left atrial appendage opening, ADD, and PDD measured by 3D-TEE were positively correlated with the size of device (r_s=0.596, 0.683, 0.606, 0.725, 0.788, 0.670, 0.823, and 0.836, respectively, P<0.001). When using PDD as the independent variable and the average diameter of the device after release the dependent variable, the following linear regression equation was obtained: the average diameter of the device after release = 0.663PDD+6.337.

Preoperative TEE measurement of the inner diameter of the left atrial appendage opening has very important guiding value in the selection of the size of the Watchman left atrial appendage closure device. Especially, PDD obtained by three-dimensional TEE has the best correlation with the device size, and may be more helpful to guide the size selection of Watchman left atrial appendage closure device before the procedure.

To measure preoperative right ventricular function in patients with coronary artery bypass grafting (CABG) by automatic functional imaging (AFI), and to evaluate its value in predicting the risk of short-term adverse events after CABG.

A total of 104 patients who received CABG at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology and Wuhan Central Hospital from December 2019 to December 2021 were selected to measure left and right ventricular cardiac function parameters by routine echocardiography and AFI, respectively. The patients were divided into an adverse event group (38 cases) or a no adverse event group (66 cases) according to whether adverse events occurred within 30 days after surgery. General data, conventional echocardiographic parameters, and AFI parameters were compared between the two groups. Receiver operating characteristic (ROC) curve analysis was performed to assess the predictive efficiency of ultrasonic parameters for adverse events. Binary logistic regression was used to identify the independent risk factors affecting the prognosis of patients with CABG.

Left ventricular ejection fraction (LVEF), left ventricular global longitudinal strain (LVGLS), right ventricular free wall strain (RVFWLS), and four-chamber right ventricular longitudinal strain (RV4CLS) in the adverse event group were lower than those of the no adverse event group (P<0.05). Right ventricular myocardial function index (RVMPI) in the adverse event group was higher than that of the group without adverse event (P<0.05). ROC curve analysis showed that the area under the ROC curve of RVFWLS was 0.77, with a sensitivity of 74% and specificity of 76% (P<0.05); the area under the ROC curve of RV4CLS was 0.69, with a sensitivity of 61% and specificity of 71% (P<0.05) . Binary logistic regression analysis showed that RVFWLS (odds ratio [OR]: 8.8, P<0.01), RV4CLS (OR: 3.8, P<0.01), LVGLS (OR: 2.5, P=0.04),RVMPI (OR: 4.5, P<0.01), chronic kidney disease (OR: 4.2, P<0.01) , NYHA grade (OR: 3.0, P=0.01), and EuroScore (OR: 2.3, P=0.04) were independent risk factors for the prognosis of patients with CABG.

Automatic myocardial function imaging of the right ventricle can provide rapid, accurate, and reliable parametric information to predict short-term adverse events after CABG. RVFWLS and RV4CLS have good predictive value for patient prognosis. At the same time, clinical indicator parameters cannot be ignored as risk factors for short-term adverse events in CABG patients.

To explore the feasibility and accuracy of deep learning-based intelligent assessment of the severity of tricuspid regurgitation by color Doppler echocardiographic flow spectrometry of the tricuspid valve.

A total of 2629 patients with tricuspid regurgitation diagnosed by echocardiography at Sichuan Provincial People's Hospital from October 2022 to April 2023 were selected, and 2629 color Doppler echocardiographic flow spectrometry images of the tricuspid valve were obtained. Two senior specialists used LabelMe software to label each complete cycle, and divided them into either a training set or a validation set at a ratio of 9:1. The YOLOv5 deep learning network was used to automatically annotate the regurgitation cardiac cycle. The performance of the automatic annotation model for the cardiac cycle was evaluated using average precision, accuracy, and recall. Two senior specialists classified the severity of regurgitation into mild, moderate, and severe levels for each single cardiac cycle regurgitation spectrometry image, and created datasets. The datasets were divided into a training set, a validation set, and a test set at a ratio of 8:1:1. The ConvNeXt deep learning network was used to evaluate the severity of regurgitation. The diagnostic performance of the regurgitation severity assessment model was evaluated using accuracy, recall, specificity, and F1 score.

The model's average precision for intelligent detection of regurgitation heart cycle was 0.979, with an accuracy of 0.951 and recall rate of 0.972. The weighted average precision of the prediction results of the model's intelligent assessment of the severity of regurgitation was 0.938, with a recall rate of 0.928, specificity of 0.964, and F1 score of 0.953, all within a high value range.

The evaluation results of the intelligent assessment method for the severity of tricuspid regurgitation in echocardiography based on deep learning are highly consistent with those manually assessed by ultrasonographers, demonstrating good feasibility and diagnostic accuracy.

To propose a deep learning-based view recognition model, SlowFast-Echo, for the automatic view recognition of two-dimensional (2D) transthoracic echocardiographic dynamic images.

From August to December 2022, 722 patients who underwent 2D transthoracic echocardiography at the Department of Ultrasound Medicine, Affiliated Hospital of Medical School, Nanjing University (9 types of clinically commonly used views [including apical two-chamber, apical three-chamber, and apical four-chamber views], with a total of 2243 dynamic images) were selected, and the images of each view were divided into training set, validation set, and test set in a ratio of 5:2:3. After training and validation of the SlowFast-Echo model, the performance of the model was evaluated quantitatively in terms of accuracy, precision, recall, and F1 score, qualitatively in terms of the interpretability of the model with regard to class activation mapping, and practically in terms of the performance of the model after field deployment to the ultrasound medicine department.

The overall accuracy, precision, recall, and F1 score of the SlowFast-Echo model for dynamic image view recognition in the test set were 0.9866, 0.9847, 0.9872, and 0.9859, respectively, and the significance heatmap indicated that the model's regions of interest were generally consistent with those drawn by the physicians; e.g., the model accurately pinpointed the significant aorta and aortic valve in parasternal short axis view of great vessel (PSAX_GV) view, mitral valve in parasternal short axis view of left ventricle at mitral value level (PSAX_MV) view, and papillary muscles in parasternal short axis view of left ventricle at papillary muscle level (PSAX_PM) view. The overall accuracy, precision, recall, and F1 score of the model for view recognition after deployment were 0.9903, 0.9865, 0.9868, and 0.9865, respectively, and the average inference time on RTX 3060 GPU for a single dynamic image was (303.2±119.3) ms, which basically meets the clinical demand for immediate processing after image acquisition.

The SlowFast-Echo model proposed in this study has good performance in view recognition of 2D transthoracic echocardiographic dynamic images and inference in real time, which is practically useful.

To construct a carotid plaque ultrasound image dataset, build a dual-input BCNN-ResNet classification deep learning model for screening carotid plaque stability, and explore the efficacy of the dual-input bilinear convolutional neural network with residual network as the backbone network (BCNN-ResNet) model for automatic classification and diagnosis of carotid plaque stability.

A total of 493 cases were collected from January 2021 to March 2023 from those who underwent carotid ultrasonography at the Shanghai Eighth People's Hospital and Xinhua Hospital of Dalian University. Carotid plaque images were observed by four ultrasonographers, and 352 ultrasound images of stable carotid plaques and 691 images of vulnerable plaques were selected after comprehensive evaluation to construct a dataset containing a total of 1043 ultrasound images of carotid arteries. Using the ResNet-50 model as the base model, with the first ResNet-50 network inputting the structural plaque images to extract the structural features and the second ResNet-50 network inputting the cropped plaque images to obtain the pixel features, a dual-input BCNN model was constructed by fusing the two sets of features. By training the images with classification supervised learning, internal validation, and external validation, the newly constructed dual-input BCNN-ResNet model was compared with ResNet-34, ResNet-50, ResNet-101, single-input BCNN-ResNet-34, dual-input BCNN-ResNet-34, and single-input BCNN-ResNet-50 for diagnostic efficacy in classifying carotid plaque stability. The diagnostic efficacy of the model was assessed by applying metrics such as the area under the receiver operating characteristic (ROC) curve (AUC).

ROC curve analysis showed that the AUC of the dual-input BCNN-ResNet-50 model for the classification and diagnosis of carotid plaque stability on ultrasound images was 0.896, while the AUC values of the single-input BCNN-ResNet-50 model, ResNet-34, ResNet-50, ResNet-101, single input BCNN-ResNet-34, and dual-input BCNN-ResNet-34 model AUC were 0.878, 0.857, 0.860, 0.859, 0.864, and 0.868, respectively. The efficacy of the dual-input BCNN-ResNet-50 model was significantly better than that of the other models in classifying and diagnosing carotid plaque stability.

The dual-input BCNN-ResNet model can automatically screen ultrasound images for carotid plaque stability, and this algorithm outperforms previous diagnostic models, providing a technical reference for clinical carotid plaque stability screening.

To construct a nomogram based on ultrasonic features and clinicopathological indicators and to explore its predictive value for axillary lymph node metastasis in patients with T1-2 breast cancer.

A total of 354 patients with histopathologically confirmed T1-2 breast cancer admitted to Xijing Hospital from January 2021 to September 2022 were included. According to whether there was axillary lymph node metastasis, the patients were divided into a metastatic group of 125 cases and a non-metastatic group of 229 cases. Univariate and multivariate Logistic regression analyses were used to screen independent predictors, and the axillary ultrasound model and comprehensive model (axillary ultrasound features + breast ultrasound features + clinicopathological indicators) were constructed. Receiver operating characteristic (ROC) curves were plotted to evaluate the predictive efficiency of the models, and the predictive efficiency was compared by the Delong's test. A nomogram of the comprehensive model was plotted, and the goodness of fit, calibration, and clinical utility of the model were evaluated by the Hosmer-Lemeshow test as well as calibration curve and decision curve analyses.

The ratio of long diameter to short diameter of lymph nodes, morphological typing of lymph nodes, maximum tumor diameter, architectural distortion, body mass index, histological grade, and estrogen receptor status were identified to be independent predictors of axillary lymph node metastasis (P<0.05 for all). The areas under the ROC curves of the axillary ultrasound model and the comprehensive model were 0.741 (0.684-0.758) and 0.812 (0.767-0.858), respectively. The prediction efficiency of the comprehensive model was greater than that of the axillary ultrasound model (Z=3.5472, P<0.001).

The nomogram developed based on axillary ultrasonic features combined with clinicopathological indicators of breast cancer can improve the diagnostic efficiency for axillary lymph nodes metastasis, and provide effective reference for the staging, prognosis, and treatment of breast cancer.

To assess the value of right heart time interval measured by echocardiography in evaluating right heart function in fetuses with right ventricular outflow tract obstruction (RVOTO).

This prospective study included 28 fetuses diagnosed with RVOTO by echocardiography (RVOTO group) at Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, and 114 fetuses matched for gestational age with no abnormalities (normal group). Right heart function parameters of all the fetuses were measured, including tricuspid annular plane systolic excursion (TAPSE). Tissue Doppler imaging (TDI) was used to obtain the tricuspid anular velocity curve and to measure the peak velocity (E') in the early diastolic period and the peak velocity (A') in the late diastolic period, and E'/A' ratio was subsequently calculated. Isovolumic contraction time (ICT), isovolumic relaxation time (IRT), filling time (FT), ejection time (ET), and cardiac cycle (CT) were also measured. The myocardial performance index (RVMPI), systolic-to-diastolic time index (SDI), systolic to diastolic duration ratio (SDR), filling time fraction (FTF), and ejection time fraction (ETF) of the fetal right heart were calculated by the formulas. The parameters between the two groups were compared.

The mean values of IRT [(43.21±4.88) ms vs (45.71±5.12) ms, P<0.05], ICT [(42.79±5.59) ms vs (47.82±7.09) ms, P<0.05], (ICT+IRT) [(87.29±7.78) ms vs (93.51±10.48) ms, P<0.05], and FT value [(124.46±4.70) ms vs (142.00±15.77) ms, P<0.05] were significantly shorter in the RVOTO group than in the normal group. The RVMPI (0.52±0.06 vs 0.48±0.07, P<0.001), SDI (2.17±0.18 vs 1.95±0.24, P<0.001), SDR (1.35±0.12 vs 1.22±0.12, P<0.001), and FTF (0.31± 0.02 vs 0.34± 0.03, P<0.001) were significantly higher in the RVOTO group than in the normal group. The values of E' (6.38±1.82 vs 5.59±1.96, P<0.001) and E'/A' ratio (0.62±0.11 vs 0.51±0.12, P<0.001) of the tricuspid annulus were significantly higher in the normal group than in the RVOTO group.

The fetal echocardiographic parameters right ventricular MPI, SDI, SDR, FTF, E' value of tricuspid valve ring, and E'/A' ratio are helpful for comprehensive evaluation of right heart function in fetuses with RVOTO, and have potential clinical value in evaluating fetuses' current disease status and prognosis.

To assess the value of echocardiography in the diagnosis of isolated fetal pulmonary valvular stenosis (PVS).

From January 2019 to December 2022, 35 fetuses diagnosed with PVS by prenatal echocardiography at the Department of Ultrasound of Tianjin Central Obstetrics and Gynecology Hospital and the Department of Ultrasound of Southern District of Chengde Medical College were retrospectively analyzed. Fetal echocardiography was performed to obtain four chamber sections, left ventricular and right ventricular outflow tract sections, and three vessel sections. The transverse diameter of each heart cavity was measured, and the opening and closing of the atrioventricular valve and regurgitation were observed. All fetuses were examined by echocardiography within 24 h after delivery to measure pulmonary artery velocity and pulmonary artery cross-valve pressure difference and evaluate the degree of pulmonary artery stenosis. The results of prenatal and postnatal echocardiography were analyzed, and Pearson correlation was used to analyze the correlation between prenatal PVS grouping (forward group and reverse group) and postpartum PVS grading (mild, moderate, and severe).

Of the 35 fetuses with PVS included, 8 underwent induced labor and 27 were born. There were 15 cases in the forward prenatal ductus arteriae (DA) group and 20 cases in the reverse prenatal DA group, and there were statistically significant differences in pulmonary artery velocity, dilation of the pulmonary artery after stenosis, and tricuspid regurgitation between the two groups (P<0.05). There was no significant difference in pulmonary artery diameter, left and right pulmonary artery diameter, pulmonary valve opening, and right ventricular diameter/left ventricular diameter ratio (P>0.05). All 27 cases of PVS in neonatal period were confirmed by echocardiography, of which 15 were mild, 8 were moderate, and 4 were severe. There was a correlation between different groups of prenatal PVS and different grades of postpartum PVS (Pearson correlation coefficient=0.58, P<0.05). Mild PVS (12/14 85.71%) was found mainly in the positive prenatal group, while moderate and severe PVS (10/13 76.92%) was found predominantly in the reverse group.

Pulmonary artery velocity is fast and tricuspid valve regurgitation is severe in fetuses with isolated PVS. The possibility of moderate or higher PVS is high after DA reverse perfusion. Dilation after stenosis is an important indirect sign of fetal PVS. Fetal PVS alone can be diagnosed before birth, and echocardiography is an important means to diagnose isolated fetal PVS and estimate its postpartum grade.

To compare the effectiveness of online teaching and traditional teaching in theoretical training of vascular ultrasound among ultrasound residents.

Twenty-one first-year residents from the ultrasound training base of Peking Union Medical College Hospital in 2020 were included and randomly divided into a traditional teaching group (11 participants) and an online teaching group (10 participants). The traditional teaching group received on-site theoretical lectures, while the online teaching group received live streaming lectures. Before and after the training, all participants underwent a vascular ultrasound knowledge test and a course satisfaction survey. The test scores and course satisfaction survey results were compared between the two groups.

Before the training, the average score of the vascular ultrasound test for the traditional teaching group was (73.50±18.11) points, and for the online teaching group it was (73.18±13.09) points, with no statistically significant difference between the two groups (P=0.803). After the training, the average scores of the traditional teaching group vs the online teaching group were (87.50±10.87) points vs (85.00±10.25) points, with no significant difference between them (P=0.532). There was a significant improvement in the vascular ultrasound knowledge test scores for both groups after the training (P<0.05 for both groups). There was no statistically significant difference in the satisfaction to course content and teaching methods between the two groups (P>0.05 for both).

Online teaching in theoretical training of vascular ultrasound can achieve teaching goals and enhance the diagnostic ability of residents in vascular ultrasound, and has comparable teaching effectiveness to traditional teaching, with the advantages of convenience and online interaction.