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.

Echocardiographic assessment of pediatric heart transplantation: A single-center experience in China.

BACKGROUND: Pediatric heart transplant (HT) has become the standard of care for end-stage heart failure in children worldwide. Serial echocardiographic evaluations of graft anatomy and function during follow-up are crucial for post-HT management. However, evolution of cardiac structure and function after pediatric HT has not been well described, especially during first year post-HT. This study aimed to characterize the evolution of cardiac structure and function after pediatric HT and investigate the correlation between biventricular function with adverse clinical outcomes. METHODS: A single-center retrospective study of echocardiographic data obtained among 99 pediatric HT patients was conducted. Comprehensive echocardiographic examination was performed in all patients at 1-, 3-, 6-, 9- and 12-months post-HT. We obtained structural, functional and hemodynamic parameters from both left- and right-side heart, such as left ventricular stroke volume (LVSV), left ventricular ejection fraction (LVEF), right ventricular fractional area change (RVFAC), etc. The cardiac evolution of pediatric HT patients during first post-HT year was described and compared between different time points. We also explored the correlation between cardiac function and major adverse transplant events (MATEs). RESULTS: 1) Evolution of left heart parameters: left atrial length, mitral E velocity, E/A ratio, LVSV and LVEF significantly increased while mitral A velocity significantly decreased over the first year after HT (P < .05). Compared with 1 month after HT, interventricular septum (IVS) and left ventricular posterior wall (LVPW) decreased at 3 months but increased afterwards. (2) Evolution of right heart parameters: right ventricular base diameter and mid-diameter; right ventricular length diameter, tricuspid E velocity, E/A ratio, tricuspid annular velocity e' at free wall, and RVFAC increased, while tricuspid A velocity decreased over the first year after HT (P < .05). (3) Univariate logistic regression model suggests that biventricular function parameters at 1-year post-HT (LVEF, RVFAC, tricuspid annular plane systolic excursion and tricuspid lateral annular systolic velocity) were associated with MATEs. CONCLUSION: Gradual improvement of LV and RV function was seen in pediatric HT patients within the first year. Biventricular function parameters associated with MATEs. The results of this study pave way for designing larger and longer follow-up of this population, potentially aiming at using multiparameter echocardiographic prediction of adverse events.

Left Heart Chamber Volumetric Assessment by Automated Three-Dimensional Echocardiography in Heart Transplant Recipients.

Background: Recently, a new automated software (Heart Model) was developed to obtain three-dimensional (3D) left heart chamber volumes. The aim of this study was to verify the feasibility and accuracy of the automated 3D echocardiographic algorithm in heart transplant (HTx) patients. Conventional manual 3D transthoracic echocardiographic (TTE) tracings and cardiac magnetic resonance (CMR) images were used as a reference for comparison. Methods: This study enrolled 103 healthy HTx patients prospectively. In protocol 1, left ventricular end-diastolic volume (LVEDV), LV end-systolic volume (LVESV), left atrial max volume (LAVmax), LA minimum volume (LAVmin) and LV ejection fraction (LVEF) were obtained using the automated 3D echocardiography (3DE) and compared with corresponding values obtained through the manual 3DE. In protocol 2, 28 patients' automated 3DE measurements were compared with CMR reference values. The impacts of contour edit and surgical technique were also tested. Results: Heart Model was feasible in 97.1% of the data sets. In protocol 1, there was strong correlation between 3DE and manual 3DE for all the parameters (r = 0.77 to 0.96, p<0.01). Compared to values obtained through manual measurements, LV volumes and LVEF were overestimated by the automated algorithm and LA volumes were underestimated. All the biases were small except for that of LAVmin. After contour adjustment, the biases reduced and all the limits of agreement were clinically acceptable. In protocol 2, the correlations for LV and LA volumes were strong between automated 3DE with contour edit and CMR (r = 0.74 to 0.93, p<0.01) but correlation for LVEF remained moderate (r = 0.65, p < 0.01). Automated 3DE overestimated LV volumes but underestimated LVEF and LA volumes compared with CMR. The limits of agreement were clinically acceptable only for LVEDV and LAVmax. Conclusion: Simultaneous quantification of left heart volumes and LVEF with the automated Heart Model program is rapid, feasible and to a great degree it is accurate in HTx recipients. Nevertheless, only LVEDV and LAVmax measured by automated 3DE with contour edit seem applicable for clinical practice when compared with CMR. Automated 3DE for HTx recipients is a worthy attempt, though further verification and optimization are needed.

Assessment of biatrial function in clinically well pediatric bicaval heart transplantation patients by three-dimensional echocardiography.

Atrial size and function are closely correlated with atrial contributions to cardiovascular performance. Therefore, in this study, we aimed to assess atrial size and function in pediatric heart transplantation (HTx) patients using three-dimensional echocardiography (3DE). We enrolled 33 clinically well pediatric HTx patients and 33 healthy controls with a similar distribution of sex and age to the HTx patients. All patients underwent two-dimensional echocardiography (2DE) and 3DE. 2DE- and 3DE-derived biatrial maximal volume (Vmax), minimal volume (Vmin), ejection volume (EV), ejection fraction (EF), volume before atrial contraciton (VpreA), passive EV, passive EF, active EV and active EF were obtained in all patients. The 3D left atrail (LA) Vmax, Vmin and VpreA increased significantly in HTx patients after being indexed by BSA, while 3D LAEV and passive EV decreased significantly (P < 0.05). Moreover, the 3D LAEF, LA passive EF, and LA active EF all decreased significantly in HTx patients (P < 0.05). The 3D right atrial (RA) Vmax, Vmin, and VpreA increased significantly in HTx patients (P < 0.05), while the 3D RAEF and RA passive EF decreased significantly in HTx patients (P < 0.05). 3DE-derived LAVmax, LAVpreA, LA passive EV, LAEF, and LA passive EF were all lower than the corresponding 2D parameters. 3DE-derived RAVpreA, RA passive EV and RAEF were all lower than the corresponding 2D parameters. Atrial sizes and function assessed by 3DE- and 2DE-derived parameters, yield significantly discordant results in pediatric HTx patients. 3DE confirms significantly enlarged atrial sizes and decreased atrial functions in pediatric HTx patients.

[Morphological assessment of sulfur mustard-induced acute lung injury in rats through different routes].

OBJECTIVE: To establish an animal model of sulfur mustard (SM)-induced acute lung injury in rats through different routes and compare the morphological changes in lung tissue and cells. METHODS: One hundred and thirty-six male rats were selected and randomly divided into 5 groups, namely peritoneal cavity SM group (n=32), trachea SM group (n=32), peritoneal cavity propylene glycol group (n=32), trachea propylene glycol group (n=32), and normal control group (n=8). The rats in peritoneal cavity SM group were injected intraperitoneally with diluted SM (0.1 ml, 8 mg/kg), and the rats in trachea SM group were injected intratracheally with diluted SM (0.1 ml, 2 mg/kg). Once the rats were sacrificed at 6, 24, 48, and 72 h after SM treatment, morphological changes in lung tissue and cells were observed by light and electron microscopy. RESULTS: In the peritoneal cavity SM group, the epithelial cells of bronchioles maintained intact with increased exudate and bleeding in alveolar cavity and large areas of pulmonary consolidation under the light microscope. In the tracheal SM group, focal ulcer formed in the epithelial cells of bronchioles with increased exudate and bleeding in alveolar cavity, partial pulmonary consolidation, and compensatory emphysema in peripheral alveolar space under the light microscope. The alveolar interval areas were widened obviously in both groups in a time-dependent manner. Under the electron microscope, we observed local loss of cellular membrane in type I alveolar epithelium, broken or lost microvilli in cells of typeⅡalveolar epithelium and fuzzy mitochondrial crista as well as the appearance of ribosome detached from rough endoplasmic reticulum in both two groups. Compared with those in the trachea SM group and the control group, the ratio of the alveolar septum average area to the visual field area in the peritoneal cavity SM group at 6, 24, 48, and 72 h was significantly higher (P<0.05). CONCLUSION: The lung tissue injury through the intraperitoneal route is more severe than that through the tracheal route, while focal ulceration of bronchioles epithelial cells appears in the case of tracheal route. The degree of injury increases over time in both groups, and the cellular damage is approximately the same in both groups.