Assessment of left ventricular ejection fraction in artificial intelligence based on left ventricular opacification.

Background: Left ventricular opacification (LVO) improves the accuracy of left ventricular ejection fraction (LVEF) by enhancing the visualization of the endocardium. Manual delineation of the endocardium by sonographers has observer variability. Artificial intelligence (AI) has the potential to improve the reproducibility of LVO to assess LVEF. Objectives: The aim was to develop an AI model and evaluate the feasibility and reproducibility of LVO in the assessment of LVEF. Methods: This retrospective study included 1305 echocardiography of 797 patients who had LVO at the Department of Ultrasound Medicine, Union Hospital, Huazhong University of Science and Technology from 2013 to 2021. The AI model was developed by 5-fold cross validation. The validation datasets included 50 patients prospectively collected in our center and 42 patients retrospectively collected in the external institution. To evaluate the differences between LV function determined by AI and sonographers, the median absolute error (MAE), spearman correlation coefficient, and intraclass correlation coefficient (ICC) were calculated. Results: In LVO, the MAE of LVEF between AI and manual measurements was 2.6% in the development cohort, 2.5% in the internal validation cohort, and 2.7% in the external validation cohort. Compared with two-dimensional echocardiography (2DE), the left ventricular (LV) volumes and LVEF of LVO measured by AI correlated significantly with manual measurements. AI model provided excellent reliability for the LV parameters of LVO (ICC > 0.95). Conclusions: AI-assisted LVO enables more accurate identification of the LV endocardium and reduces observer variability, providing a more reliable way for assessing LV function.

Biventricular longitudinal strain as a predictor of functional improvement after D-shant device implantation in patients with heart failure.

Background: The creation of an atrial shunt is a novel approach for the management of heart failure (HF), and there is a need for advanced methods for detection of cardiac function response to an interatrial shunt device. Ventricular longitudinal strain is a more sensitive marker of cardiac function than conventional echocardiographic parameters, but data on the value of longitudinal strain as a predictor of improvement in cardiac function after implantation of an interatrial shunt device are scarce. We aimed to investigate the exploratory efficacy of the D-Shant device for interatrial shunting in treating heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF), and to explore the predictive value of biventricular longitudinal strain for functional improvement in such patients. Methods: A total of 34 patients were enrolled (25 with HFrEF and 9 with HFpEF). All patients underwent conventional echocardiography and two-dimensional speckle tracking echocardiogram (2D-STE) at baseline and 6 months after implantation of a D-Shant device (WeiKe Medical Inc., WuHan, CN). Left ventricular global longitudinal strain (LVGLS) and right ventricular free wall longitudinal strain (RVFWLS) were evaluated by 2D-STE. Results: The D-Shant device was successfully implanted in all cases without periprocedural mortality. At 6-month follow-up, an improvement in New York Heart Association (NYHA) functional class was observed in 20 of 28 patients with HF. Compared with baseline, patients with HFrEF showed significant reduced left atrial volume index (LAVI) and increased right atrial (RA) dimensions, as well as improved LVGLS and RVFWLS, at 6-month follow-up. Despite reduction in LAVI and increase in RA dimensions, improvements in biventricular longitudinal strain did not occur in HFpEF patients. Multivariate logistic regression demonstrated that LVGLS [odds ratio (OR): 5.930; 95% CI: 1.463-24.038; P = 0.013] and RVFWLS (OR: 4.852; 95% CI: 1.372-17.159; P = 0.014) were predictive of improvement in NYHA functional class after D-Shant device implantation. Conclusion: Improvements in clinical and functional status are observed in patients with HF 6 months after implantation of a D-Shant device. Preoperative biventricular longitudinal strain is predictive of improvement in NYHA functional class and may be helpful to identify patients who will experience better outcomes following implantation of an interatrial shunt device.

Echocardiography of a left ventricular apical thrombus after myocardial infarction: A complementary role of the linear transducer array.

We presented a 28-year-old man with myocardial infarction with an apical mass detected by echocardiography. Compared with routine imaging using a phased transducer array, linear transducer arrays were used in two-dimensional and contrast echocardiography to enhance apical visualization and successfully demonstrated the shape and blood perfusion of apical thrombus with precision. We successfully explored the feasibility of imaging the apical thrombus with a linear transducer array.

Automatic view classification of contrast and non-contrast echocardiography.

Background: Contrast and non-contrast echocardiography are crucial for cardiovascular diagnoses and treatments. Correct view classification is a foundational step for the analysis of cardiac structure and function. View classification from all sequences of a patient is laborious and depends heavily on the sonographer's experience. In addition, the intra-view variability and the inter-view similarity increase the difficulty in identifying critical views in contrast and non-contrast echocardiography. This study aims to develop a deep residual convolutional neural network (CNN) to automatically identify multiple views of contrast and non-contrast echocardiography, including parasternal left ventricular short axis, apical two, three, and four-chamber views. Methods: The study retrospectively analyzed a cohort of 855 patients who had undergone left ventricular opacification at the Department of Ultrasound Medicine, Wuhan Union Medical College Hospital from 2013 to 2021, including 70.3% men and 29.7% women aged from 41 to 62 (median age, 53). All datasets were preprocessed to remove sensitive information and 10 frames with equivalent intervals were sampled from each of the original videos. The number of frames in the training, validation, and test datasets were, respectively, 19,370, 2,370, and 2,620 from 9 views, corresponding to 688, 84, and 83 patients. We presented the CNN model to classify echocardiographic views with an initial learning rate of 0.001, and a batch size of 4 for 30 epochs. The learning rate was decayed by a factor of 0.9 per epoch. Results: On the test dataset, the overall classification accuracy is 99.1 and 99.5% for contrast and non-contrast echocardiographic views. The average precision, recall, specificity, and F1 score are 96.9, 96.9, 100, and 96.9% for the 9 echocardiographic views. Conclusions: This study highlights the potential of CNN in the view classification of echocardiograms with and without contrast. It shows promise in improving the workflow of clinical analysis of echocardiography.

Artificial Intelligence-Enhanced Echocardiography for Systolic Function Assessment.

The accurate assessment of left ventricular systolic function is crucial in the diagnosis and treatment of cardiovascular diseases. Left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) are the most critical indexes of cardiac systolic function. Echocardiography has become the mainstay of cardiac imaging for measuring LVEF and GLS because it is non-invasive, radiation-free, and allows for bedside operation and real-time processing. However, the human assessment of cardiac function depends on the sonographer's experience, and despite their years of training, inter-observer variability exists. In addition, GLS requires post-processing, which is time consuming and shows variability across different devices. Researchers have turned to artificial intelligence (AI) to address these challenges. The powerful learning capabilities of AI enable feature extraction, which helps to achieve accurate identification of cardiac structures and reliable estimation of the ventricular volume and myocardial motion. Hence, the automatic output of systolic function indexes can be achieved based on echocardiographic images. This review attempts to thoroughly explain the latest progress of AI in assessing left ventricular systolic function and differential diagnosis of heart diseases by echocardiography and discusses the challenges and promises of this new field.

Transesophageal echocardiography in detecting a dislodged coronary stent.

We report a rare case of coronary stent dislodgement after percutaneous coronary intervention. This case demonstrates the value of two-dimensional and threedimensional transesophageal echocardiography in detecting the dislodged stent at the aortic root and may be beneficial in management.

Clinical Analysis of Risk Factors for Mortality in Type A Acute Aortic Dissection: A Single Study From China.

Objective: Acute type A aortic dissection (ATAAD) is a fatal condition that requires emergency surgery. The aim of the present study was to determine pre- and intra-operative risk factors for in-hospital mortality in patients with ATAAD. Methods: Consecutive 313 patients with ATAAD who underwent emergency surgery at our hospital from February 2012 to February 2017 were enrolled in our study. Univariate and multivariate logistic regression analysis were performed to identify the pre-operative and intra-operative risk factors for in-hospital mortality. Results: Of the 313 patients, 32 patients (10.2%) died. Compared with survivors, non-survivors had higher heart rate, serum potassium level and EuroSCORE II, and higher incidence of moderate to severe pericardial effusion, supra-aortic vessels involvement, myocardial ischemia and lower-extremity ischemia. As for surgery-related factors, the duration of surgery and cardiopulmonary bypass time were longer in non-survivors than survivors. In addition, non-survivors were more likely to undergo coronary-artery bypass graft compared with survivors. On multivariate analysis, elevated plasma potassium level (OR: 43.0, 95% CI: 3.8-51.5, p < 0.001), high incidence of supra-aortic vessels involvement (OR: 4.4, 95% CI: 1.5-7.0, p = 0.008) and lower-extremity ischemia (OR: 4.9, 95% CI: 1.6-6.9; p = 0.009), and longer duration of surgery (OR 6.0, 95% CI: 1.8-18.7, p = 0.000) and cardiopulmonary bypass time (OR: 3.7, 95% CI: 1.3-9.3, p = 0.001) were independently predictive of higher mortality in patients with ATAAD. Conclusions: Supra-aortic vessels involvement, lower-extremity ischemia and elevated plasma potassium level are independent predictors of mortality in patients with ATAAD. A significant decrease in duration of surgery and cardiopulmonary bypass time is helpful to improve survival of patients.

Right renal angiomyolipoma with intracardiac extension and pulmonary embolism.

Angiomyolipoma is the most common benign mesenchymal neoplasm of the kidney. Despite its benign nature, it may, in rare instances, invade the inferior vena cava and right cardiac chambers. A tumor thrombus in the pulmonary artery is an extremely rare but potentially fatal complication. We report a rare case of 53-year-old woman showing a right renal angiomyolipoma with intracardiac extension and pulmonary embolism. The patient underwent open radical right nephrectomy and thrombectomy of the inferior vena cava and pulmonary artery. Our case highlights that close attention to these complications is important to determine the surgical plan.

Prognostic Value of Right Ventricular Two-Dimensional and Three-Dimensional Speckle-Tracking Strain in Pulmonary Arterial Hypertension: Superiority of Longitudinal Strain over Circumferential and Radial Strain.

BACKGROUND: Right ventricular (RV) dysfunction is a predictor of adverse outcomes in patients with pulmonary arterial hypertension (PAH). Three-dimensional (3D) speckle-tracking echocardiography (STE) has been increasingly used to quantify RV function. However, the strain parameters evaluated by two-dimensional (2D) STE and 3D STE, which provide the most valuable clinical information, remain unknown. The purpose of our study was to investigate whether RV longitudinal strain (LS) provided a superior estimation of RV systolic performance and prognostic information compared with other strain vectors. METHODS: We prospectively studied 54 treatment-naïve patients with PAH and 35 normal controls. Pulmonary artery systolic pressure classified patients with PAH into three subgroups. Patients with PAH underwent echocardiography, cardiac magnetic resonance (CMR) imaging, 6-minute walking tests, and right-sided cardiac catheterization before and six months after vasodilator therapy. The 2D LS, 3D LS, circumferential strain (CS), and radial strain (RS) of RV free wall were calculated by 2D and 3D STE. RV ejection fraction (RVEF) was obtained from CMR. The patients were followed for a predefined endpoint of PAH-related hospitalization and death. RESULTS: Our findings revealed that 2D and 3D LS showed significant reduction in mild PAH patients, whereas CS and RS were decreased in moderate and severe PAH patients. Right ventricular 3D LS had a similar correlation with CMR RVEF and hemodynamic parameters as 2D LS and the other strain vectors. The 2D and 3D LS improved 6 months after vasodilator therapy (P < .001 for both). After a median follow-up of 28 months, 20 patients had endpoint events. Receiver operating characteristic curve analysis demonstrated that RV 3D LS displayed a similar diagnostic performance for detecting adverse cardiac events as 2D LS (area under the curve: 0.84 vs 0.76, P = .11). Separate multivariable Cox analysis showed that RV 2D LS (hazard ratio [HR] = 1.19; 95% CI, 1.03～1.45; P = .01) and 3D LS (HR = 1.28; 95% CI, 1.08～1.52; P = .005) were significant predictors of adverse outcomes. CONCLUSIONS: Patients with PAH show reduced RV strain. Two-dimensional and 3D LS can track clinical improvement following vasodilator therapy and provide valuable prognostic information.

Relationship between the abnormal diastolic vortex structure and impaired left ventricle filling in patients with hyperthyroidism.

Intraventricular hydrodynamics plays an important role in evaluating cardiac function. Relationship between diastolic vortex and left ventricular (LV) filling is still rarely elucidated. The aim of this study was to evaluate the evolution of vortex during diastole in hyperthyroidism (HT) and explore the alteration of hydromechanics characteristics with sensitive indexes.Forty-three patients diagnosed with HT were classified into 2 groups according to whether myocardial damage existed: simple hyperthyroid group (HT1, n = 21) and thyrotoxic cardiomyopathy (HT2, n = 22). Twenty-seven age- and gender-matched healthy volunteers were enrolled as the control group. Offline vector flow mapping (VFM model) was used to analyze the LV diastolic blood flow patterns and fluid dynamics. Hemodynamic parameters, vortex area (A), circulation (C), and intraventricular pressure gradient (ΔP), in different diastolic phases (early, mid, and late) were calculated and analyzed.HT2, with a lower E/A ratio and left ventricular ejection fraction (LVEF), had a larger left atrium diameter (LAD) compared with those of the control group and HT1 (P < .05). Compared with the control group, the vortex size and strength, intraventricular pressure gradient during early and mid-diastole were higher in HT1 and lower in HT2 (P < .05). And in late diastole, the vortex size and strength, intraventricular pressure gradient of HT2 became higher than those of the control group (P < .05). Good correlation could be found between CE and E/A (P < .05), CM and ΔPM (P < .01), CL and FT3 (P < .05).VFM is proven practical for detecting the relationship between the changes of left ventricular diastolic vortex and the abnormal left ventricular filling.

Left ventricular systolic intraventricular flow field assessment in hyperthyroidism patients using vector flow mapping.

Intraventricular hydrodynamics is considered an important component of cardiac function assessment. Vector flow mapping (VFM) is a novel flow visualization method to describe cardiac pathophysiological condition. This study examined use of new VFM and flow field for assessment of left ventricular (LV) systolic hemodynamics in patients with simple hyperthyroidism (HT). Thirty-seven simple HT patients were enrolled as HT group, and 38 gender- and age-matched healthy volunteers as control group. VFM model was used to analyze LV flow field at LV apical long-axis view. The following flow parameters were measured, including peak systolic velocity (Vs), peak systolic flow (Fs), total systolic negative flow (SQ) in LV basal, middle and apical level, velocity gradient from the apex to the aortic valve (ΔV), and velocity according to half distance (V1/2). The velocity vector in the LV cavity, stream line and vortex distribution in the two groups were observed. The results showed that there were no significant differences in the conventional parameters such as left ventricular ejection fraction (LVEF), left ventricular end-diastolic diameter (LVEDD) and left atrium diameter (LAD) between HT group and control group (P>0.05). Compared with the control group, a brighter flow and more vortexes were detected in HT group. Non-uniform distribution occurred in the LV flow field, and the stream lines were discontinuous in HT group. The values of Vs and Fs in three levels, SQ in middle and basal levels, ΔV and V1/2 were higher in HT group than in control group (P<0.01). ΔV was positively correlated with serum free thyroxin (FT4) (r=0.48, P<0.01). Stepwise multiple regression analysis showed that LVEDD, FT4, and body surface area (BSA) were the influence factors of ΔV. The unstable left ventricular systolic hydrodynamics increased in a compensatory manner in simple HT patients. The present study indicated that VFM may be used for early detection of abnormal ventricle contraction in clinical settings.

Wave intensity analysis of carotid artery: a noninvasive technique for assessing hemodynamic changes of hyperthyroid patients.

This study examined the wave intensity (WI) of the carotid artery in patients with hyperthyroid in order to assess the hemodynamic changes of hyperthyroid patients. A total of 86 hyperthyroid patients without cardiac morphological changes and arrhythmia, and 80 healthy control subjects were enrolled in the study. Right common carotid artery (RCCA) was selected for ultrasonic imaging to obtain WI indices, including amplitude of the peak during early systole (W1), amplitude of the peak during late systole (W2), area of the negative wave during mid-systole (NA), interval between R wave of electrocardiogram and W1 (R-1st), interval between W1 and W2 (1st-2nd). The levels of serum thyroid hormones, consisting of free triiodothyronine (FT3), free thyroxin (FT4) and thyroid stimulating hormone (TSH), were measured in hyperthyroid patients. Echocardiographic indices including left ventricular ejection fraction (LVEF) and left ventricular fraction shortening (LVFS) were determined in each subject. The results showed that the W1, W2, NA, and (1st-2nd×HR) in hyperthyroid patients were significantly higher than those in healthy controls. There was no significant difference in LVEF and LVFS between the two groups. FT3 was correlated with W1, W2, NA, (1st-2nd×HR), pulse pressure (PP) and heart rate (HR) in hyperthyroid patients. Several abnormal waves on WI curves were present in 19 hyperthyroid patients during mid-systole. It was concluded that WI technique may prove a real-time, noninvasive, sensitive and convenient tool for assessing the cardiac function and hemodynamic alterations in hyperthyroid patients.

Wave Intensity Analysis of Carotid Artery: A Noninvasive Technique for Assessing Hemodynamic Changes of Hyperthyroid Patients / 华中科技大学学报（医学）（英德文版）

This study examined the wave intensity (WI) of the carotid artery in patients with hyperthyroid in order to assess the hemodynamic changes of hyperthyroid patients. A total of 86 hyperthyroid patients without cardiac morphological changes and arrhythmia, and 80 healthy control subjects were enrolled in the study. Right common carotid artery (RCCA) was selected for ultrasonic imaging to obtain WI indices, including amplitude of the peak during early systole (W1), amplitude of the peak during late systole (W2), area of the negative wave during mid-systole (NA), interval between R wave of electrocardiogram and W1 (R-1 st), interval between W1 and W2 (1 st-2nd). The levels of serum thyroid hormones, consisting of free triiodothyronine (FT3), free thyroxin (FT4) and thyroid stimulating hormone (TSH), were measured in hyperthyroid patients. Echocardiographic indices including left ventricular ejection fraction (LVEF) and left ventricular fraction shortening (LVFS) were determined in each subject. The results showed that the WI, W2, NA, and (lst-2nd×HR) in hyperthyroid patients were significantly higher than those in healthy controls. There was no significant difference in LVEF and LVFS between the two groups. FT3 was correlated with WI, W2, NA,(lst-2nd×HR), pulse pressure (PP) and heart rate (HR) in hyperthyroid patients. Several abnormal waves on WI curves were present in 19 hyperthyroid patients during mid-systole. It was concluded that WI technique may prove a real-time, noninvasive, sensitive and convenient tool for assessing the cardiac function and hemodynamic alterations in hyperthyroid patients.