BEAS-Net: a Shape-Prior-Based Deep Convolutional Neural Network for Robust Left Ventricular Segmentation in 2D Echocardiography.

Left ventricle (LV) segmentation of 2D echocardiography images is an essential step in the analysis of cardiac morphology and function and - more generally - diagnosis of cardiovascular diseases. Several deep learning (DL) algorithms have recently been proposed for the automatic segmentation of the LV, showing significant performance improvement over the traditional segmentation algorithms. However, unlike the traditional methods, prior information about the segmentation problem, e.g. anatomical shape information, is not usually incorporated for training the DL algorithms. This can degrade the generalization performance of the DL models on unseen images if their characteristics are somewhat different from those of the training images, e.g. low-quality testing images. In this study, a new shape-constrained deep convolutional neural network (CNN) - called BEAS-Net - is introduced for automatic LV segmentation. The BEAS-Net learns how to associate the image features, encoded by its convolutional layers, with anatomical shape-prior information derived by the B-spline explicit active surface (BEAS) algorithm to generate physiologically meaningful segmentation contours when dealing with artifactual or low-quality images. The performance of the proposed network was evaluated using three different in-vivo datasets and was compared a deep segmentation algorithm based on the U-Net model. Both networks yielded comparable results when tested on images of acceptable quality, but the BEAS-Net outperformed the benchmark DL model on artifactual and low-quality images.

A patient-specific echogenic soft robotic left ventricle embedded into a closed-loop cardiovascular simulator for advanced device testing.

Cardiovascular medical devices undergo a large number of pre- and post-market tests before their approval for clinical practice use. Sophisticated cardiovascular simulators can significantly expedite the evaluation process by providing a safe and controlled environment and representing clinically relevant case scenarios. The complex nature of the cardiovascular system affected by severe pathologies and the inherently intricate patient-device interaction creates a need for high-fidelity test benches able to reproduce intra- and inter-patient variability of disease states. Therefore, we propose an innovative cardiovascular simulator that combines in silico and in vitro modeling techniques with a soft robotic left ventricle. The simulator leverages patient-specific and echogenic soft robotic phantoms used to recreate the intracardiac pressure and volume waveforms, combined with an in silico lumped parameter model of the remaining cardiovascular system. Three different patient-specific profiles were recreated, to assess the capability of the simulator to represent a variety of working conditions and mechanical properties of the left ventricle. The simulator is shown to provide a realistic physiological and anatomical representation thanks to the use of soft robotics combined with in silico modeling. This tool proves valuable for optimizing and validating medical devices and delineating specific indications and boundary conditions.

Low-Dose Dobutamine Stress Echocardiography for the Early Detection of Pulmonary Arterial Hypertension in Selected Patients with Systemic Sclerosis Whose Resting Echocardiography Is Non-Diagnostic for Pulmonary Hypertension.

BACKGROUND: Dobutamine stress echocardiography (DSE) has limited application in systemic sclerosis (SSc). We examined DSE usefulness in revealing pulmonary arterial hypertension (PAH) in selected SSc patients whose resting echocardiography for pulmonary hypertension (PH) was non-diagnostic. METHODS: Forty SSc patients underwent right heart catheterization (RHC) and, simultaneously, low-dose DSE (incremental doses up to 20 µg/kg/min). Inclusion criteria were: preserved left and right ventricular (RV) function (tricuspid annulus plane systolic excursion [TAPSE] ≥ 16 mm and tissue Doppler imaging-derived systolic velocity of tricuspid annulus [RVS'] > 10 cm/s), normal pulmonary function tests, and baseline maximal tricuspid regurgitation (TR) velocity of 2.7-3.2 m/s. RESULTS: Of 36 patients who completed DSE, resting RHC diagnosed PAH in 12 patients (33.3%). At 20 µg/kg/min, patients with PAH had higher TR velocity, higher pulmonary arterial pressure measured by RHC, and lower RV inotropic response compared with patients without PAH. A cut-off value of maximal TR velocity >3.1 m/s had a sensitivity of 80%, a specificity of 84.2%, and an accuracy of 82.4% for the detection of PAH. CONCLUSIONS: Low-dose DSE has a satisfactory diagnostic accuracy for the early detection of PAH in highly selected SSc patients whose baseline echocardiographic measurements for PH lie in the gray zone.

The Role of Exercise Doppler Echocardiography to Unmask Pulmonary Arterial Hypertension in Selected Patients with Systemic Sclerosis and Equivocal Baseline Echocardiographic Values for Pulmonary Hypertension.

Recently, a lower mean pulmonary arterial pressure (PAP) cutoff of >20 mmHg for pulmonary hypertension (PH) definition has been proposed. We examined whether exercise Doppler echocardiography (EDE) can unmask PA hypertension (PAH) in systemic sclerosis (SSc) patients whose baseline echocardiography for PH is equivocal. We enrolled 49 patients with SSc who underwent treadmill EDE. Tricuspid regurgitation (TR) velocity was recorded immediately after EDE. Inotropic reserve of right ventricle (RV) was assessed by the change (post-prior to exercise) of tissue Doppler imaging-derived peak systolic velocity (S) of tricuspid annulus. Inclusion criteria comprised preserved left and RV function, and baseline TR velocity between 2.7 and 3.2 m/s. All patients had right-heart catheterization (RHC) within 48 h after EDE. From 46 patients with good quality of post-exercise TR velocity, RHC confirmed PAH in 21 (45.6%). Post-exercise TR velocity >3.4 m/s had a sensitivity of 90.5%, a specificity of 80% and an accuracy of 84.8% in detecting PAH. Inotropic reserve of RV was positively correlated with maximum achieved workload in METs (r = 0.571, p < 0.001). EDE has a good diagnostic accuracy for the identification of PAH in selected SSc patients whose baseline echocardiographic measurements for PH lie in the gray zone, and it is also potentially useful in assessing RV contractile reserve.

Echocardiography in Autoimmune Rheumatic Diseases for Diagnosis and Prognosis of Cardiovascular Complications.

Autoimmune rheumatic diseases are systemic diseases frequently affecting the heart and vessels. The main cardiovascular complications are pericarditis, myocarditis, valvular disease, obstructive coronary artery disease and coronary microcirculatory dysfunction, cardiac failure and pulmonary hypertension. Echocardiography, including transthoracic two and three-dimensional echocardiography, Doppler imaging, myocardial deformation and transesophageal echo, is an established and widely available imaging technique for the identification of cardiovascular manifestations that are crucial for prognosis in rheumatic diseases. Echocardiography is also important for monitoring the impact of drug treatment on cardiac function, coronary microcirculatory function, valvular function and pulmonary artery pressures. In this article we summarize established and evolving knowledge on the role of echocardiography for diagnosis and prognosis of cardiovascular abnormalities in rheumatic diseases.