Noninvasive Staging of Hepatic Steatosis Using Calibrated 2D US with Liver Biopsy as the Reference Standard.

Background Accumulation of lipid in the liver (ie, hepatic steatosis) is the basis of nonalcoholic fatty liver disease (NAFLD). Asymptomatic steatosis can lead to nonalcoholic steatohepatitis and downstream complications. Purpose To assess the diagnostic performance of calibrated US (CAUS) as a method for detection and staging of hepatic steatosis in comparison with liver biopsy. Materials and Methods Two-dimensional US images in 223 consecutive patients who underwent US-guided liver biopsy from May 2012 to February 2016 were retrospectively analyzed by two observers using CAUS. CAUS semiautomatically estimates echo-level and texture parameters, with particular interest in the residual attenuation coefficient (RAC), which is the remaining steatosis-driven attenuation obtained after correction of the beam profile. Data were correlated with patient characteristics and histologically determined steatosis grades and fibrosis stages. The data were equally divided into training and test sets to independently train and test logistic regression models for detection (>5% fat) and staging (>33% and >66% fat) of hepatic steatosis by using area under the receiver operating characteristic curve (AUC) analysis. Results A total of 195 patients (mean age, 50 years ± 13 [SD]; 110 men) were included and divided into a training set (n = 97 [50%]) and a test set (n = 98 [50%]). The average CAUS interobserver correlation coefficient was 0.95 (R range, 0.87-0.99). The best correlation with steatosis was found for the RAC parameter (R = 0.78, P < .01), while no correlation was found for fibrosis (R = 0.14, P = .054). Steatosis detection using RAC showed an AUC of 0.97 (95% CI: 0.94, 1.00), and the multivariable AUC was found to be 0.97 (95% CI: 0.95, 1.00). The predictive performance for moderate and severe hepatic steatosis using RAC was 0.93 (95% CI: 0.88, 0.98) and 0.93 (95% CI: 0.87, 0.98), respectively. Conclusion The calibrated US parameter residual attenuation coefficient detects and stages steatosis accurately with limited interobserver variability, and performance is not hampered by the presence of fibrosis. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Grant in this issue.

Modeling envelope statistics of blood and myocardium for segmentation of echocardiographic images.

The objective of this study was to investigate the use of speckle statistics as a preprocessing step for segmentation of the myocardium in echocardiographic images. Three-dimensional (3D) and biplane image sequences of the left ventricle of two healthy children and one dog (beagle) were acquired. Pixel-based speckle statistics of manually segmented blood and myocardial regions were investigated by fitting various probability density functions (pdf). The statistics of heart muscle and blood could both be optimally modeled by a K-pdf or Gamma-pdf (Kolmogorov-Smirnov goodness-of-fit test). Scale and shape parameters of both distributions could differentiate between blood and myocardium. Local estimation of these parameters was used to obtain parametric images, where window size was related to speckle size (5 x 2 speckles). Moment-based and maximum-likelihood estimators were used. Scale parameters were still able to differentiate blood from myocardium; however, smoothing of edges of anatomical structures occurred. Estimation of the shape parameter required a larger window size, leading to unacceptable blurring. Using these parameters as an input for segmentation resulted in unreliable segmentation. Adaptive mean squares filtering was then introduced using the moment-based scale parameter (sigma(2)/mu) of the Gamma-pdf to automatically steer the two-dimensional (2D) local filtering process. This method adequately preserved sharpness of the edges. In conclusion, a trade-off between preservation of sharpness of edges and goodness-of-fit when estimating local shape and scale parameters is evident for parametric images. For this reason, adaptive filtering outperforms parametric imaging for the segmentation of echocardiographic images.

Objective performance testing and quality assurance of medical ultrasound equipment.

There is an urgent need for a measurement protocol and software analysis for objective testing of the imaging performance of medical ultrasound equipment from a user's point of view. Methods for testing of imaging performance were developed. Simple test objects were used, which have a long life expectancy. First, the elevational focus (slice thickness) of the transducer was estimated and the in-plane transmit focus was positioned at the same depth. Next, the postprocessing look-up-table (LUT) was measured and linearized. The tests performed were echo level dynamic range (dB), contrast resolution (i.e., gamma of display, number of gray levels/dB) and sensitivity, overall system sensitivity, lateral sensitivity profile, dead zone, spatial resolution and geometric conformity of display. The concept of a computational observer was used to define the lesion signal-to-noise ratio, SNR(L) (or Mahalanobis distance), as a measure for contrast sensitivity. All the measurements were made using digitized images and quantified by objective means, i.e., by image analysis. The whole performance measurement protocol, as well as the quantitative measurements, have been implemented in software. An extensive data-base browser was implemented from which analysis of the images can be started and reports generated. These reports contain all the information about the measurements, such as graphs, images and numbers. The approach of calibrating the gamma by using a linearized LUT was validated by processing simultaneously acquired rf data. The contrast resolution and echo level of the rf data had to be compressed by a factor of two and amplified by a gain factor corresponding to 12 dB. This resulted in contrast curves that were practically identical to those obtained from DICOM image data. The effects of changing the transducer center frequency on the spatial resolution and contrast sensitivity were estimated to illustrate the practical usefulness of the developed approach of quality assurance by measuring objective performance characteristics. The developed methods might be considered as a minimum set of objective quality assurance measures. This set might be used to predict clinical performance of medical ultrasound equipment, taking into account the performance at a unique point in space i.e., the coinciding depths of the elevation and in-plane (azimuth) foci. Furthermore, it should be investigated whether the approach might be used to compare objectively various brands of equipment and to evaluate the performance specifications given by the manufacturer. Last but not least, the developed approach can be used to monitor, in a hospital environment, the medical ultrasound equipment during its life cycle. The software package may be viewed and downloaded at the website http://www.qa4us.eu.

Segmentation of the heart muscle in 3-D pediatric echocardiographic images.

This study aimed to show segmentation of the heart muscle in pediatric echocardiographic images as a preprocessing step for tissue analysis. Transthoracic image sequences (2-D and 3-D volume data, both derived in radiofrequency format, directly after beam forming) were registered in real time from four healthy children over three heart cycles. Three preprocessing methods, based on adaptive filtering, were used to reduce the speckle noise for optimizing the distinction between blood and myocardium, while preserving the sharpness of edges between anatomical structures. The filtering kernel size was linked to the local speckle size and the speckle noise characteristics were considered to define the optimal filter in one of the methods. The filtered 2-D images were thresholded automatically as a first step of segmentation of the endocardial wall. The final segmentation step was achieved by applying a deformable contour algorithm. This segmentation of each 2-D image of the 3-D+time (i.e., 4-D) datasets was related to that of the neighboring images in both time and space. By thus incorporating spatial and temporal information of 3-D ultrasound image sequences, an automated method using image statistics was developed to perform 3-D segmentation of the heart muscle.

Three-Dimensional Model-Based Segmentation in Echocardiography Using High Temporal Tissue and Blood Flow Information.

Accurate 3-D surface segmentation is a challenging task in echocardiography because of the relatively low image quality. We introduce a new method for 3-D segmentation of the endocardium involving temporal decorrelation of echo signals originating from tissue and blood using radiofrequency (RF) signals acquired in 3-D Doppler mode. Temporal features were extracted in 3-D Doppler mode, where a sequence of RF lines is recorded for each image line. Each set of RF lines is highly correlated because of the high pulse repetition frequency. However, for high blood flow, the RF signals will decorrelate over time in contrast to the endocardium, which will remain relatively highly correlated over time. These decorrelation features permit differentiation between myocardial tissue and blood flow. We describe an implementation of a 3-D segmentation model in which temporal information is used as external constraint. The model was validated in a phantom and in vivo in healthy volunteers (n = 5). The phantom study revealed that the model successfully segmented the artificial blood lumen even for low flow velocity and illustrated the sensitivity of the segmentations to flow rate. In healthy volunteers, high Dice similarity indices indicate that 3-D segmentation of the endocardial border in vivo is feasible.

Improvement of heart function after balloon dilation of congenital valvar aortic stenosis: a pilot study with ultrasound tissue Doppler and strain rate imaging.

The aim was to investigate the effects of balloon dilation of congenital valvar aortic (Ao) stenosis on heart function with conventional and with new echocardiographic techniques. Nine patients, preballoon and 1 to 4 d postballoon dilation of Ao-valve, were included in the study. Assessment of heart function was made by using conventional echo/Doppler, tissue Doppler imaging (TDI) and strain rate imaging (SRI). Mean (and standard deviation) of posttreatment drop of aortic valve pressure gradient was 34.1 (sd 14.0) mmHg, p < 0.01. Conventional echo/Doppler end-diastolic left ventricular posterior wall (LVPW) thickness and interventricular septum (IVS) thickness did not change significantly. Mean change of LV fractional shortening (FS) was -5.2 (sd 3.2)%, p < 0.01. The observed changes of FS did not significantly correlate to the magnitude of pressure gradient changes. Changes of TDI and SRI parameters indicated that an increase in absolute value is observed in most cases, but correlation to pressure gradient change remains poor, with a few exceptions, both in LV free wall (LVFW) and IVS. Data from IVS are more consistent than of LVFW. It is concluded that the global functional parameter FS assessed by conventional echo/Doppler has diagnostic value for the assessment of (improved) heart function already shortly after intervention, when compared with the pretreatment value. Local parameters from the new echographic techniques show less significant short-term effects attributable to the intervention. Improvement of the precision of SRI measurements is needed. A larger study is indicated to fully investigate the expected potentials of TDI and SRI for the assessment of local improvement of heart function early after intervention, as well as for revealing eventual late effects on these functional parameters.

Quantitative Ultrasound for Staging of Hepatic Steatosis in Patients on Home Parenteral Nutrition Validated with Magnetic Resonance Spectroscopy: A Feasibility Study.

Patients on home parenteral nutrition are at risk for developing liver dysfunction, which is due partly to the accumulation of lipids in the liver (steatosis) and may progress to end-stage liver disease with overt liver failure. Therefore, a timely diagnosis with easy access to repeated assessment of the degree of liver steatosis is of great importance. A pilot study was performed in 14 patients on long-term home parenteral nutrition using the computer-aided ultrasound method. Ultrasound radio frequency data were acquired using a phased array transducer and were converted into conventional B-mode images. All patients were subjected to proton magnetic resonance spectroscopy measurement of liver fat content for reference. Computer-aided ultrasound parameters similar to those in a previous validation study in cows revealed significant correlations with fat content measured by magnetic resonance spectroscopy. The most significant parameters were the residual attenuation coefficient (R = 0.95, p < 0.001) and the lateral speckle size (R = 0.77, p = 0.021). These findings indicate the potential usefulness of computer-aided ultrasound for staging of hepatic steatosis.

Automated Assessment of Right Ventricular Volumes and Function Using Three-Dimensional Transesophageal Echocardiography.

Assessment of right ventricular (RV) function is known to be of diagnostic value in patients with RV dysfunction. Because of its complex anatomic shape, automated determination of the RV volume is difficult and strong reliance on geometric assumptions is not desired. A method for automated RV assessment was developed using three-dimensional (3-D) echocardiography without relying on a priori knowledge of the cardiac anatomy. A 3-D adaptive filtering technique that optimizes the discrimination between blood and myocardium was applied to facilitate endocardial border detection. Filtered image data were incorporated in a segmentation model to automatically detect the endocardial RV border. End-systolic and end-diastolic RV volumes, as well as ejection fraction, were computed from the automatically segmented endocardial surfaces and compared against reference volumes manually delineated by two expert cardiologists. The results reported good performance in terms of correlation and agreement with the results from the reference volumes.

Effects of skin on bias and reproducibility of near-infrared spectroscopy measurement of cerebral oxygenation changes in porcine brain.

The influence of skin on the bias and reproducibility of regional cerebral oxygenation measurements is investigated using cw near-infrared spectroscopy (NIRS). Receiving optodes are placed over the left and right hemispheres of a piglet (C3, C4 EEG placement code) and one transmitting optode centrally (Cz position). Optical densities (OD) are measured during stable normo, mild, and deep hypoxemia. This is done for skin condition 1: all optodes on the skin; skin condition 2: transmitting optode on the skin and one receiving optode on the skull; and skin condition 3: all optodes on the skull. Absolute changes of oxy- (cO2Hb), deoxyhemoglobin (cHHb), and total hemoglobin (ctHb) concentrations [micromolL] are calculated from the ODs. These absolute changes are calculated for each skin condition with respect to normoxic condition. Additionally, for skin condition 2, the difference of concentration changes between receiver 1 (skull) and receiver 2 (skin) is calculated. The effect of skin removal is an average increase of attenuation changes by a factor of 1.66 (=0.51 OD) and of the concentration changes due to the arterial oxygen saturation steps by 23%. We conclude that skin significantly influences regional oxygenation measurements. Nevertheless, it is hypothesized that the estimated concentration changes are dominated by changes of the oxygenation in the brain.

Modeling ultrasound contrast measurement of blood flow and perfusion in biological tissue.

The aim of this paper was to obtain a parametric description of ultrasound (US) contrast dilution curves obtained in piglet brain after a bolus injection of contrast agent (Sonovue). Several models for so-called dye-dilution curves were considered, and a choice for the gamma probability density function (pdf) was made. This function was chosen primarily because of the elegant mathematical opportunities to achieve a practical one-parameter curve fitting and to extend this model for incorporating recirculation. The model was fitted to US contrast data obtained in piglet brain. The software enables interactive insertion of curve landmarks and a single parameter (i.e., order of gamma pdf) iterative least squares fit was then made to the first bolus passage part of the curve. Sequential recirculation curve fits were calculated from this fit. A preliminary result for 24 experimental curves is a median value of 3 (range 3 to 5) for the order of the gamma function (kappa) and a median of 0.6/s (range 0.3 to 0.9) for the event rate (lambda). The theoretical background for obtaining the relative plasma flow and volume in tissue (i.e., "blood flow index", BFI, and "blood volume index", BVI, respectively) from the reconstructed first passage curve is given as well. The potential of assessment of the relative fraction of left-to-right shunting of the cardiac output in case of persistent ductus arteriosis is outlined. The requirements for an absolute measurement of tissue blood flow and tissue perfusion are discussed. The conclusion is that the gamma model can well be fitted to the experimental contrast curves; thus, yielding the parameters kappa and lambda. Assessment of BFI and BVI is feasible.

Assessment of local changes of cerebral perfusion and blood concentration by near infrared spectroscopy and ultrasound contrast densitometry.

The objective of this study is to correlate regional cerebral blood concentration measurements made with near infrared spectroscopy to simultaneous local measurements of ultrasound contrast agent (CA) densitometry. Experiments were performed with piglets (7 kg) under general anesthesia. The cerebral blood flow (CBF) and volume (CBV) were changed by inducing various degrees of hypercapnia. NIRS measurements were performed with a quasi-continuous wave system, using an optode distance of 3-6 cm. The concentration changes in oxygenated and deoxygenated hemoglobin and their sum and difference (cO2Hb, cHHb, ctHb, cHbD) were continuously calculated. Ultrasound contrast agent (SF6) was administered as a short intra-venous bolus. Ultrasound equipment was used in pulse inversion second harmonic gray scale imaging mode at low transmit power setting. Three regions-of-interest (0.25 cm2) were analyzed in each image. Wash-in curves were constructed as spatial mean gray level vs. time. The variables collected with both methods changed according to the induced changes in the physiological condition. Changes in the PaCO2, pH and carotid flow induced highly correlated changes in cO2Hb, cHHb, ctHb and cHbD, and in the variables derived from CA analyses. NIRS and CA methods measure regional, respectively, local changes in CBV and CBF. Moreover, NIRS can yield complementary information about the cerebral oxygenation.

Photoacoustic determination of blood vessel diameter.

A double-ring sensor was applied in photoacoustic tomographic imaging of artificial blood vessels as well as blood vessels in a rabbit ear. The peak-to-peak time (tau(pp)) of the laser (1064 nm) induced pressure transient was used to estimate the axial vessel diameter. Comparison with the actual vessel diameter showed that the diameter could be approximated by 2ctau(pp), with c the speed of sound in blood. Using this relation, the lateral diameter could also precisely be determined. In vivo imaging and monitoring of changes in vessel diameters was feasible. Finally, acoustic time traces were recorded while flushing a vessel in the rabbit ear with saline, which proved that the main contribution to the laser-induced pressure transient is caused by blood inside the vessel and that the vessel wall gives only a minor contribution.

Left ventricular myocardial function in congenital valvar aortic stenosis assessed by ultrasound tissue-velocity and strain-rate techniques.

A pilot study was performed to reveal the potentials of new echo Doppler techniques for the detection of myocardial changes due to congenital valvar aortic stenosis. A total of 24 patients, (age range 0.1 to 17 years), with various degrees of aortic stenosis, and 24 age- and gender-matched, healthy children were enrolled in this study. Conventional echo Doppler, tissue velocity imaging (TVI) and strain-rate imaging (StRI) measurements were carried out using the apical four-chamber view and transthoracic long-axis view. All patients had normal fractional shortening of the left ventricle (> 28%). Although the sum of septal and ventricular wall thicknesses was significantly increased in the patients (p < 0.001), only 6 of the 24 patients showed left ventricular hypertrophy. In tissue velocity mode, systolic and early diastolic wall velocity acceleration was significantly reduced in both views. Peak systolic and early diastolic wall velocities, as well as strain rate values, in the four-chamber view were significantly reduced in the patient group. The decrease was highest for the strain-rate values in all cases. In conclusion, strain rate values at different moments within the heart cycle might become important parameters in the assessment of myocardial impairment. Further studies are indicated to assess the correlation of these parameters with the severity of stenosis, left ventricular hypertrophy and irreversible myocardial function changes.

Cardiological Ultrasound Imaging.

This review paper is intended for the interested outsider of the field of echocardiography and it presents a short introduction into the numerous ultrasound (US) methods and techniques for anatomical and functional diagnosis of the heart. The basic techniques are generally used for some times already, as there are one dimensional (1D) M(otion) mode, the real time 2D B(rightness) mode technique and the various Doppler measurement techniques and imaging modes. The M-mode technique shows the movements of the tissue in a 1D B-mode display vs. time. The 2D B-mode images are showing the heart contractions and dilations in real time, thus making this technique the basic tool for detecting anatomical disturbances and myocardial (localized) abnormal functioning. Improved image quality is achieved by Second Harmonic Imaging and myocardial perfusion can be quantified using Contrast Agent Imaging. Doppler techniques were introduced in the fifties of last century and used for blood flow velocity measurement. Continuous wave (CW) Doppler has the advantage of allowing measurement of high velocities, as may occur in vascular or valvular stenosis and insufficiency. The exact location of the major Doppler signal received cannot be estimated making this technique ambiguous in some clinical problems. Single gated Pulse Wave (PW) Doppler velocity measurement delivers exact location of the measurement position by using an interactively positioned time (=depth) gate in which the velocity is being measured. The disadvantage of this technique is the relatively low maximum velocity that can be measured. Multigate PW Doppler techniques can be used for the assessment of a velocity profile over the vessel cross section. A more sophisticated use of this technique is the combination with 2D B-mode imaging in the color Doppler mode, called "color flow mapping", in which the multigate Doppler signal is color coded and shown in 2D format overlayed in the conventional 2D B mode image. In the past two decades, technique to quantify myocardial function were developed: Tissue Doppler Imaging (TDI), Strain Rate and Strain Imaging. The temporal resolution of ultrasound imaging can be further improved by Plane Wave Imaging, and Synthetic Aperture Imaging. The recent introduction of 2D matrix transducers extended the real time imaging potential by allowing 3D imaging and sophisticated segmentation techniques for the estimation of quantitative functional parameters, as for instance cardiac output.

Enhancing the performance of lateral shear strain estimation by using 2-D strain imaging.

Radio-frequency (RF) ultrasound can be used to estimate deformation of biological tissue. Decorrelation of sequentially acquired ultrasound signals resulting from the deformation imposes a limitation on the precision (elastographic signal-to-noise ratio; SNRe) of estimating these deformations; this is presented as the lateral shear strain filter. In this paper, we explore the effect of a 2-D-window-based strain estimation approach on the lateral shear strain filter and propose an extension of the 1-D theoretical lateral shear strain filter to 2-D. We compared the performance of the 2-D approach in simulated ultrasound data and a tissue-mimicking phantom with that of the 2-D lateral shear strain filter. In simulations, the 2-D-window-based approach shows an effect in the axial direction similar to the 2-D prediction. In simulations and experiments, increasing the window size in the lateral direction shows an increase in the maximum SNRe of the lateral shear strain filter. Increasing the lateral overlap has no effect on the estimation of lateral shear strain. These results were confirmed in the tissue-mimicking phantom experiments. When compared with the 2-D lateral shear strain filter, the results obtained with the 2-D-window-based approach showed an enhanced performance by incorporating the lateral window size in the lateral shear strain estimation, which was consistent with the proposed theory.

New developments in paediatric cardiac functional ultrasound imaging.

Ultrasound imaging can be used to estimate the morphology as well as the motion and deformation of tissues. If the interrogated tissue is actively deforming, this deformation is directly related to its function and quantification of this deformation is normally referred as 'strain imaging'. Tissue can also be deformed by applying an internal or external force and the resulting, induced deformation is a function of the mechanical tissue characteristics. In combination with the load applied, these strain maps can be used to estimate or reconstruct the mechanical properties of tissue. This technique was named 'elastography' by Ophir et al. in 1991. Elastography can be used for atherosclerotic plaque characterisation, while the contractility of the heart or skeletal muscles can be assessed with strain imaging. Rather than using the conventional video format (DICOM) image information, radio frequency (RF)-based ultrasound methods enable estimation of the deformation at higher resolution and with higher precision than commercial methods using Doppler (tissue Doppler imaging) or video image data (2D speckle tracking methods). However, the improvement in accuracy is mainly achieved when measuring strain along the ultrasound beam direction, so it has to be considered a 1D technique. Recently, this method has been extended to multiple directions and precision further improved by using spatial compounding of data acquired at multiple beam steered angles. Using similar techniques, the blood velocity and flow can be determined. RF-based techniques are also beneficial for automated segmentation of the ventricular cavities. In this paper, new developments in different techniques of quantifying cardiac function by strain imaging, automated segmentation, and methods of performing blood flow imaging are reviewed and their application in paediatric cardiology is discussed.

Persistent reduction in left ventricular strain using two-dimensional speckle-tracking echocardiography after balloon valvuloplasty in children with congenital valvular aortic stenosis.

BACKGROUND: The aim of this study was to investige serial changes of myocardial deformation using two-dimensional speckle-tracking echocardiographic (2DSTE) imaging in children undergoing balloon valvuloplasty for congenital valvular aortic stenosis (VAS). METHODS: Thirty-seven children with isolated congenital VAS were enrolled in this study prospectively. Patients underwent echocardiographic evaluation at three instances: before balloon valvuloplasty, 6 months after intervention, and 3 years after intervention. Longitudinal, circumferential, and radial peak systolic strain values were determined, as well as systolic strain rate and the time to peak global systolic strain. Linear mixed statistical models were used to assess changes in 2DSTE parameters after balloon intervention. Using one-way analysis of variance, 2DSTE results at 3-year follow-up were compared with 2DSTE measurements in 74 healthy age-matched children and 76 children with uncorrected VAS whose severity of stenosis corresponded to residual stenosis of study subjects at 3-year follow-up. RESULTS: Global peak strain and strain rate measurements in all three directions were decreased before intervention compared with healthy children. Global peak strain and strain rate measurements increased significantly (P < .001) several months after balloon valvuloplasty and continued to increase at 3-year follow-up. However, at 3-year follow-up, global peak strain and strain rate in the longitudinal and circumferential directions were significantly lower (P < .001) compared with both control groups. Measurements of time to peak global systolic strain were significantly shorter at early follow-up compared with measurements before intervention (P < .05). CONCLUSIONS: Shortly after balloon valvuloplasty for severe congenital VAS, there is an improvement in systolic myocardial deformation. However, 2DSTE parameters do not return to normal at 3-year follow-up. These abnormalities in systolic deformation cannot be fully attributed to residual stenosis or aortic regurgitation.

Transcutaneous vs. intraoperative quantitative ultrasound for staging bovine hepatic steatosis.

The aim of this study was to test the hypothesis that quantitative analysis of transcutaneous (Transc) ultrasound (US) images can predict the liver fat content with similar accuracy and precision as using intraoperative (Intraop) US. The second goal was to investigate if a tissue mimicking phantom (TMP) might be used as reference for automatic gain compensation (AGC) vs. depth instead of using the data of a set of cows without hepatic alterations. A study was performed in post partum dairy cows (N = 151), as an animal model of human nonalcoholic fatty liver disease (NAFLD), to test these hypotheses. Five Transc and five Intraop US liver images were acquired in each animal and a liver biopsy was taken. In liver tissue samples, triacylglycerol (TAG) content was measured by biochemical analysis and hepatic alterations, other than hepatic steatosis, were excluded by clinical examination. Several preprocessing steps were performed before the ultrasound tissue characteristics (UTC) parameters of B-mode images were derived. Stepwise multiple linear regression analysis was performed on a training set (N = 76) and the results were used on the test group (N = 75) to predict the TAG content in the liver. In all cases, the residual attenuation coefficient (ResAtt) was the only selected parameter. Receiver operating characteristics (ROC) analysis was applied to assess the performance and area under the curve (AUC) of predicting TAG and to compare the sensitivity and specificity of the methods used. High ROC values for AUC (95%), sensitivity (87%) and specificity (83%) for both Intraop and Transc applications with control group as well as with phantom-based AGC were obtained. Consequently, it can be concluded that Transc results are equivalent to Intraop results. Furthermore, equivalent ROC values, when using TMP AGC, indicates the potential use of TMP-based corrections instead of normal group-based corrections. The high predictive values indicate that noninvasive quantitative US has a great potential for staging and screening on hepatic steatosis in cows.

Correlation based 3-D segmentation of the left ventricle in pediatric echocardiographic images using radio-frequency data.

Clinical diagnosis of heart disease might be substantially supported by automated segmentation of the endocardial surface in three-dimensional (3-D) echographic images. Because of the poor echogenicity contrast between blood and myocardial tissue in some regions and the inherent speckle noise, automated analysis of these images is challenging. A priori knowledge on the shape of the heart cannot always be relied on, e.g., in children with congenital heart disease, segmentation should be based on the echo features solely. The objective of this study was to investigate the merit of using temporal cross-correlation of radio-frequency (RF) data for automated segmentation of 3-D echocardiographic images. Maximum temporal cross-correlation (MCC) values were determined locally from the RF-data using an iterative 3-D technique. MCC values as well as a combination of MCC values and adaptive filtered, demodulated RF-data were used as an additional, external force in a deformable model approach to segment the endocardial surface and were tested against manually segmented surfaces. Results on 3-D full volume images (Philips, iE33) of 10 healthy children demonstrate that MCC values derived from the RF signal yield a useful parameter to distinguish between blood and myocardium in regions with low echogenicity contrast and incorporation of MCC improves the segmentation results significantly. Further investigation of the MCC over the whole cardiac cycle is required to exploit the full benefit of it for automated segmentation.

Three-dimensional cardiac strain imaging in healthy children using RF-data.

In this study, a new radio-frequency (RF)-based, three-dimensional (3-D) strain imaging technique is introduced and applied to 3-D full volume ultrasound data of the heart of healthy children. Continuing advances in performance of transducers for 3-D ultrasound imaging have boosted research on 3-D strain imaging. In general, speckle tracking techniques are used for strain imaging. RF-based strain imaging has the potential to yield better performance than speckle- based methods because of the availability of phase information but such a system output is commercially not available. Furthermore, the relatively low frame rate of 3-D ultrasound data has limited broad application of RF-based cardiac strain imaging. In this study, the previously reported two-dimensional (2-D) strain methodology was extended to the third dimension. Three-dimensional RF-data were acquired in 13 healthy children, in the age range of 6-15 years, at a relatively low frame rate of 38-51 Hz. A 3-D, free-shape, coarse-to-fine displacement and strain estimation algorithm was applied to the RF-data. The heart was segmented using 3-D ellipsoid fitting. Strain was estimated in the radial (R), circumferential (C) and longitudinal directions (L). Our preliminary results reveal the applicability of the 3-D strain estimation technique on full volume 3-D RF-data. The technique enabled 3-D strain imaging of all three strain components. The average strains for all children were in the lateral wall R = 37 ± 10% (infero-lateral) and R = 32% ± 10% (antero-lateral), C = -9% ± 4% (antero-lateral) and C = -9% ± 4% (infero-lateral), L = -18% ± 6 % (antero-lateral) and L = -15% ± 4% (infero-lateral). In the septum, strains were found to be R = 24% ± 10% (antero-septal) and R = 13% ± 5% (infero-septal), C = -13% ± 5% (antero-septal) and -13% ± 5% (infero-septal) and L = -13% ± 3% (antero-septal) and L = -16% ± 5% (infero-septal). Strain in the anterior and inferior walls seemed underestimated, probably caused by the low (in-plane) resolution and poor image quality. The field-of-view as well as image quality were not always sufficient to image the entire left ventricle. It is concluded that 3-D strain imaging using RF-data is feasible, but validation with other modalities and with conventional 3-D speckle tracking techniques will be necessary.