Narrow Results

The quality of medical images is crucial for accurately diagnosing and treating various diseases. However, current automated methods for assessing image quality are based on neural networks, which often focus solely on pixel distortion and overlook the significance of complex structures within the images. This study introduces a novel neural network model designed explicitly for automated image quality assessment that addresses pixel and semantic distortion. The model introduces an adaptive ranking mechanism enhanced with contrast sensitivity weighting to refine the detection of minor variances in similar images for pixel distortion assessment. More significantly, the model integrates a structure-aware learning module employing graph neural networks. This module is adept at deciphering the intricate relationships between an image’s semantic structure and quality. When evaluated on two ultrasound imaging datasets, the proposed method outshines existing leading models in performance. Additionally, it boasts seamless integration into clinical workflows, enabling real-time image quality assessment, crucial for precise disease diagnosis and treatment.

In this paper, we are presenting a method for achieving strain analysis of cardiac tissues from 3D ultrasound images. First of all, the left ventricle was segmented by an iterative snake algorithm which exploits the gradient vector flow field estimated from the 3D images. Once the ventricle was segmented, a set of points located on the external surface of the ventricle was selected. The movement of all these points along the entire cardiac cycle was calculated by using speckle tracking techniques. The strain of all the segments connecting two of the considered points was computed, together with the rotation of the points along the long axis of the ventricle. The method was tested on images acquired from 10 subjects: three with heart pathologies, five healthy subjects, and two patients with mild hypertrophy. Preliminary results showed the feasibility of characterizing healthy subjects and patients with well-defined heart pathologies by using the outcomes achieved with the strain analysis.

This study is aimed to assess (1) Left ventricle (LV) contractile function and ventricular-arterial matching from echocardiography; (2) whether ventricular-arterial matching (VAM) is associated with N-terminal pro B-type natriuretic peptide (NT-proBNP), and stroke output in patients with heart failure. Normal subjects (n = 81) and heart failure patients (n = 80) underwent echocardiography, Doppler echocardiography and tissue Doppler imaging. Only heart failure patients underwent blood test for NT-proBNP. The LV contractility was calculated as dσ ⁎ /dt_max = 3 × (dV/dt)_max/2V_m = 3 × V_peak × (π × D²/4)/(2V_m), and the arterial elastance was calculated as E_a = SBP × 0.9/SV, wherein V_peak and D are peak velocity and diameter of LV outflow tract, V_m is myocardial volume, SBP is the systolic blood pressure and SV is stroke volume measured from LVOT. The VAM index was expressed as the ratio of LV contractility to arterial elastance (dσ ⁎ /dt_max/E_a). We found that HF patients had (i) decreased dσ ⁎ /dt_max (1.46 ± 0.73 versus 4.06 ± 1.06 s^-1), (ii) increased E_a (2.90 ± 0.87 versus 1.81 ± 0.38 mmHg/mL), and (iii) attenuated ventricular-arterial matching index (0.66 ± 0.57 versus 2.38 ± 0.91 mL/mmHg⋅s) (all p < 0.001) compared with normal subjects. The VAM index was correlated inversely with NT-proBNP (r = -0.32, p < 0.05), but positively with the stroke volume (r = 0.85, p < 0.001). The VAM index of < 1.51 was able to clearly differentiate the failing heart from normal hearts (AUC = 0.959, Sensitivity = 0.911, Specificity = 0.905). Heart failure patients demonstrated impaired ventricular contractility, enhanced arterial stiffening, and attenuated ventricular-arterial matching index. The attenuated ventricular-arterial matching index value was associated with elevated NT-proBNP levels and lower cardiac output.

This study aims to explore the biomechanics of positive intrathoracic pressure and its effects on left ventricle (LV) filling in healthy subjects. 30 healthy subjects were enlisted to perform a Valsalva maneuver (VM) with a load of 40mmHg lasted 10s. LV filling parameters were measured by echocardiography at rest and at 10s during the VM. Compared with the at rest values, LV early inflow velocity (E) decreased significantly ($p<0.05$), late inflow velocity (A) decreased insignificantly ($p>0.05$), while the E/A ratio decreased significantly ($p<0.05$) after 10s of the strain phase of the VM. LV septal early tissue velocity ($e_{\mathrm{\text{sep}}}$) and lateral early tissue velocity ($e_{\mathrm{\text{lat}}}$) of the mitral did not change ($p>0.05$), while the $E∕e_{\mathrm{\text{sep}}}$ ratio and the $E∕e_{\mathrm{\text{lat}}}$ ratio decreased significantly ($p<0.05)$ after 10s during the VM. Based on these results, biomechanical analysis suggests that the effects of positive intrathoracic pressure on the LV free wall impede LV diastolic motion, which may be one of the factors contributing to a decrease in E and the E/A ratio. Positive intrathoracic pressure also increases the flow resistance of the LV and pulmonary vasculature, which may contribute to a decrease in E, the $E∕e_{\mathrm{\text{sep}}}$ ratio, and the $E∕e_{\mathrm{\text{lat}}}$ ratio.

Parasternal and apical echocardiography images captured from different cross-sectional planes (short-axis and four chambers) convey significant information about the structure and function of infarcted Left Ventricular (LV) myocardium. Thus, features from these cross-sectional views of echocardiograms extracted using computer-aided techniques may aid in characterizing Myocardial Infarction (MI). Therefore, this paper proposes a new algorithm for automated MI characterization using features extracted from parasternal short axis and apical four chambers cross-sectional views of 160 subjects (80 with MI and 80 normal) echocardiograms. The Stationary Wavelet Transform (SWT) method is used to extract the Relative Wavelet Energy and Entropy (RWE and RWEnt) features from the two cross-sectional views of echocardiography images separately. These features are ranked and subjected to classification in two different steps: (i) the features from each view are separately ranked using entropy, t-test and Wilcoxon ranking tests and fed to the classifier, and (ii) later, the features from both the views are combined and ranked. Finally, these ranked features are subjected to the Support Vector Machine (SVM) classifier for characterization of normal and MI using a minimum number of features. The proposed method is able to identify MI with 95.0% of accuracy, 93.7% of sensitivity and 96.2% of specificity using 32 features extracted from parasternal short-axis view; an accuracy of 96.2%, sensitivity of 97.5% and specificity of 95.0% with 18 apical four chamber view features. The results show that by combining the features from both views enables the confirmation of MI LVs with an accuracy of 96.8%, sensitivity of 93.7% and specificity of 100% using 16 features extracted from only two frames. Software development is in progress which can be incorporated into the echocardiography ultrasound machine for automated detection of MI patients.

Objective: To analyze the anatomical morphological and hemodynamic characteristics of left ventricular outflow tract stenosis (LVOTS) by echocardiography and MRI. Methods: The Echocardiography data of 112 patients with LVOTS were retrospectively analyzed by measuring the basal interventricular septal thickness (IVST-b), the left ventricle posterior wall thickness (LVPWT), and the peak pressure gradient of LVOTS, as well as by observing the characteristics of spectral pattern and the presence of systolic anterior motion of mitral valve leaflets. A Siemens 3.0T MRI scanner was used to scan the contrast-enhanced left ventricular (LV) volume of all cases. The obtained end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV) and ejection fraction (EF) of LV were compared with the Echocardiography results. Results: The 112 patients were divided into four groups: hypertrophic obstructive cardiomyopathy (Group I, 42 cases), hypertensive left ventricular hypertrophy (Group II, 40 cases), basal septal hypertrophy in the elderly (Group III, 21 cases), and the subaortic membrane (Group IV, 9 cases). The continuous wave (CW) Doppler characteristic of Groups I, II, and III was broadsword-shaped jet, and that of Group IV was equilateral triangle-like spectrum. The IVST-b, IVST-b/LVPWT ratio and peak pressure gradient of LVOTS of Group I was significantly higher than those in Groups II and III ($P<0.001$). The LVPWT of the first three groups was slightly correlated with the LVOTS peak pressure gradient ($r=0.404$, respectively, $P<0.001$). There were no statistically significant differences between Echocardiography and MRI results regarding the LV EDV, ESV, SV, and EF ($P>0.05$), and no statistically significant differences were found between Echocardiography and MRI results regarding the myocardial thicknesses of septal, anterior, lateral, and inferior walls ($P>0.05$). The Pearson’s correlation analysis demonstrated correlations between MRI and Echocardiography results for LV EDV, ESV, SV, and EF ($P=0.002$, 0.002, 0.009, and $<0.001$, respectively). The MRI enhancement was shown as delayed enhancement in 52 cases, localized enhancement in 8 cases, diffuse enhancement in 6 cases, and transmural enhancement in 3 cases, with abnormal enhancement lesions distributed in the area of ventricular septum free wall junctions and ventricular septum. Conclusion: Using MRI to evaluate LV function of hypertrophic cardiomyopathy is reliable and accurate. MRI enhancement can be used for the quantitative measurement of myocardial fibrosis. Echocardiography can distinguish the stenosis types of LVOTS. The IVST-b and existence of SAM may be important anatomical characteristics determining the degree of dynamic stenosis, and MRI combined with Echocardiography can provide important detailed information.

In this paper, an automatic method for segmentation of the left ventricle in two-dimensional (2D) echocardiography images of one cardiac cycle is proposed. In the first step of this method, using a mean image of a sequence of echocardiography images and its statistical properties the approximate region of left ventricle (LV) is extracted. Then the coordinate of extracted rectangular (ROI) is applied on all frames of sequences automatically. The mean image extracted ROI is used for defining the initial contour by scanning from the center point in polar coordinate. In the next step, all the extracted ROIs from the frames are mapped in a 2D space using the nonlinear dimension reduction manifold learning method. Using the properties of the manifold map end diastole (ED) and end systole (ES) frames are determined. Segmentation of the frames begins from ES frame, utilizing the dynamic directional vector field convolution (DDVFC) level set method with the initial contour as mentioned above. Final contour of each segmented frame is used as the initial contour of the next frame. Maximum range of the active contour motion is limited by a percent of the Euclidean distance between the point corresponds the current frame and the previous one in the resultant manifold. The results obtained from our method are quantitatively evaluated to those obtained by the gold contours drawn by a cardiologist on 489 echocardiographic images of seven volunteers using four distance measures: Hausdorff distance, average distance, area difference and area coverage error. We have also compared our results with the results of applying only DDVFC method. Comparing the implementation of only the DDVFC method, the results show final contours by proposed method are more close to contours drawn by a cardiologist.

Identifying end-diastole (ED) and end-systole (ES) frames is highly important in the process of evaluating cardiac function and measuring global parameters accurately, such as ejection fraction (EF), cardiac output (CO) and stroke volume (SV). The aim of this study is to develop a new method based on measuring volume changes in left ventricle (LV) during cardiac cycle. For this purpose, the level set method is used both in detecting endocardium border and quantifying cardiac function of all frames. Demonstrating LV volumes at a time displays ED and ES frames and volumes used in calculating required parameters. Since ES and ED frames exist in iso-volumic phases of the cardiac cycle with minimum and maximum values of LV volume signals, such peaks can be utilized in finding related frames. The results obtained from 44 volunteers, as golden values, were validated by an experienced echocardiologist through manual annotation and comparison of EF, CO and SV measurements. These measurements prove the suggested method provides accurate and reliable automatic detection of ED and ES frames.

Left ventricular (LV) shape alteration is closely correlated with cardiac disease and LV function. In this paper, we propose a feature to detect LV dysfunction globally by analyzing the LV shape deformation in systolic contraction. The feature is an index that is extracted from geometric measurement of LV shape such as the length of the long axis, the short axis, and the apical diameter. A framework for computing the features is also proposed that consists of shape model construction and motion estimation of myocardial boundary. The LV shape model is extracted from apical 2 and 4 chamber views of 2D echocardiography. The long axis, the short axis, and the apical diameter were redefined according to the LV shape constructed. An optical flow technique was used to estimate the position of the LV boundary in each frame. The classification of the LV dysfunction was performed using linear discriminant analysis (LDA) and neural networks (NNs). The 2D echocardiography dataset collected from routine clinical check-up were used to validate the proposed method by comparing the computation result and cardiac expert diagnose. Classification performance and statistical analysis, which was performed to discriminate between healthy and diseased data, indicated promising results. The global LV features would provide a strong basis for a global LV function diagnosis and a global cardiac pathology assessment.

This paper describes a novel method to noninvasively measure the myocardial viscoelasticity in vivo to evaluate the heart diastolic properties. By the ultrasonic measurement of the myocardial motion, we have already found that some pulsive waves are spontaneously excited by aortic-valve closure (AVC) at end-systole (T₀). In this study, a sparse sector scan at a sufficiently high frame rate clearly reveals wave propagation along the heart wall. The propagation time of the wave along the heart wall is very small, namely, several milliseconds, and cannot be measured by conventional equipment. From the measured phase velocity, we estimate the myocardial viscoelasticity in vivo. In in vivo experiments applied to 6 healthy subjects, the propagation of the pulsive wave was clearly visible in all subjects. For the frequency component up to 90 Hz, the typical propagation speed is about several m/s and rapidly decreased around the time of AVC. For the healthy subject, the typical value of elasticity was about 24-30 kPa and did not change around the time of AVC. The typical transient values of viscosity decreased rapidly from 400 Pa·s to 70 Pa·s around the time of AVC. The measured shear elasticity and viscosity in this study are comparable to those obtained for the human tissues using audio frequency in in vitro experiments reported in the literature. This method offers potential for in vivo imaging of the spatial distribution of the passive mechanical properties of the myocardium, which cannot be obtained by conventional echocardiography, CT, or MRI.

Color Doppler flow imaging is the most widely used technique for assessing mitral regurgitation, but eccentric regurgitant jets cannot be visualized and measured by conventional 2D techniques. A new procedure was developed for quantitative analysis of mitral eccentric regurgitation jet based on real-time three-dimensional color Doppler flow images with mitral regurgitation and 3D acquisition. The experiments display the mitral regurgitation volume tendency chart within different systolic phase and the regurgitation volumes were measured by real-time 3D Color Doppler. A higher degree of mitral regurgitation was found in the patients with eccentric jets. While traditional 2D color Doppler jet areas failed to identify patients with different grades of regurgitation, jet volumes could so discriminate. Real-time 3D Doppler revealed new patterns of regurgitant flow and allowed a more accurate semi-quantitative assessment of complex eccentric regurgitant jets. The Real-time three-dimensional color Doppler has a great potential for becoming a reference method for the assessment of eccentric regurgitant with heart valve disease.