Ultrasonic backscatter system for automated on-line endocardial boundary detection: evaluation by ultrafast computed tomography

Sensitivity and specificity of left ventricular ejection fraction by echocardiographic automated border detection: comparison with radionuclide ventriculography

BACKGROUND

Echocardiographic automated border detection (ABD) provides on-line, beat-to-beat estimation of left ventricular (LV) ejection fraction (EF). Sensitivity and specificity of using ABD-EF for diagnosing LV dysfunction in routine clinical situations have not been previously studied.

HYPOTHESIS

Analysis of ABD-EF data based on receiver operating characteristic (ROC) should provide useful information about sensitivity and specificity for clinical diagnosis of LV function based on ABD-EF.

METHODS

The study group included 50 consecutive patients with EF measured by both ABD and radionuclide ventriculography (RVG). ABD-EF was recorded for 25 consecutive heart beats in the apical four-chamber view. Data were analyzed statistically by linear regression, Bland-Altman plot, and ROC. In ROC analysis, abnormal LV function was defined RVG-EF < or = 40%.

RESULTS

ABD and RVG showed a moderate correlation in the EF measurements: slope = 0.93, intercept = 17%, r = 0.79 (n = 50). Interbeat variability in ABD was diminished by averaging consecutive beats; standard error of estimate (SEE) decreased from 15.6% without averaging to 12.5% with 25-beat averaging. Bland-Altman analysis indicated that ABD-EF compared unfavorably with RVG-EF, with limits of agreement from -11% to 39%. ABD-EF showed a systematic overestimation (p < 0.005), which was compensated by increasing the threshold for abnormal ABD-EF to 56%. With the optimized threshold, ABD-EF provided 89% sensitivity and 89% specificity (85% overall diagnostic accuracy) for diagnosing abnormal LV function.

CONCLUSION

This study explored the limitations of on-line echocardiographic measurement of EF in a clinical setting and provided useful data for assessing interbeat variability, sensitivity, and specificity.

A Novel Two-Dimensional Echocardiographic Image Analysis System Using Artificial Intelligence-Learned Pattern Recognition for Rapid Automated Ejection Fraction

OBJECTIVES

We sought to test the hypothesis that a novel 2-dimensional echocardiographic image analysis system using artificial intelligence-learned pattern recognition can rapidly and reproducibly calculate ejection fraction (EF).

BACKGROUND

Echocardiographic EF by manual tracing is time consuming, and visual assessment is inherently subjective.

METHODS

We studied 218 patients (72 female), including 165 with abnormal left ventricular (LV) function. Auto EF incorporated a database trained on >10,000 human EF tracings to automatically locate and track the LV endocardium from routine grayscale digital cineloops and calculate EF in 15 s. Auto EF results were independently compared with manually traced biplane Simpson's rule, visual EF, and magnetic resonance imaging (MRI) in a subset.

RESULTS

Auto EF was possible in 200 (92%) of consecutive patients, of which 77% were completely automated and 23% required manual editing. Auto EF correlated well with manual EF (r = 0.98; 6% limits of agreement) and required less time per patient (48 +/- 26 s vs. 102 +/- 21 s; p < 0.01). Auto EF correlated well with visual EF by expert readers (r = 0.96; p < 0.001), but interobserver variability was greater (3.4 +/- 2.9% vs. 9.8 +/- 5.7%, respectively; p < 0.001). Visual EF was less accurate by novice readers (r = 0.82; 19% limits of agreement) and improved with trainee-operated Auto EF (r = 0.96; 7% limits of agreement). Auto EF also correlated with MRI EF (n = 21) (r = 0.95; 12% limits of agreement), but underestimated absolute volumes (r = 0.95; bias of -36 +/- 27 ml overall).

CONCLUSIONS

Auto EF can automatically calculate EF similarly to results by manual biplane Simpson's rule and MRI, with less variability than visual EF, and has clinical potential.

Performance of User Independent Echocardiographic Border Detection Algorithm: Comparison with Human Observer Variability

INTRODUCTION

We evaluated a method for autonomous, user-independent automated border delineation (ABD) developed by Geiser and Wilson, by comparing the accuracy of ABD relative to manual border tracing.

METHODS

Short axis echocardiographic images of 84 patients from 3 clinical sites were analyzed using ABD and by manual tracing performed by two observers at each site and two observers at a core laboratory. The centerline method was used to measure the distance between each pair of computer-generated and hand-traced borders. Cardiac parameters were also measured from all sets of borders: LV area, fractional area change, antero-posterior diameter, wall motion, and wall thickening.

RESULTS

The distance between computer-generated and hand-traced borders was slightly but significantly greater than human interobserver variability between the clinical sites and the core laboratory (0.34+/-0.25 (N = 328) vs. 0.26+/-0.16 (N = 320) cm for the endocardium at end diastole, p = 0.0001). Measurements of LV area and fractional area change were similar by ABD and manual tracing. Other cardiac parameters showed greater deviation between ABD and manually traced borders than between human observers.

CONCLUSION

Autonomous ABD provides accurate measurements of LV area and area-derived indices. However measurements dependent on border point location deviate more by ABD.

Volumetric Quantification of Global and Regional Left Ventricular Function From Real-Time Three-Dimensional Echocardiographic Images

Background— Real-time 3D echocardiographic (RT3DE) data sets contain dynamic volumetric information on cardiac function. However, quantification of left ventricular (LV) function from 3D echocardiographic data is performed on cut-planes extracted from the 3D data sets and thus does not fully exploit the volumetric information. Accordingly, we developed a volumetric analysis technique aimed at quantification of global and regional LV function.

Methods and Results— RT3DE images obtained in 30 patients (Philips 7500) were analyzed by use of custom software based on the level-set approach for semiautomated detection of LV endocardial surface throughout the cardiac cycle, from which global and regional LV volume (LVV)–time and wall motion (WM)–time curves were obtained. The study design included 3 protocols. In protocol 1, time curves obtained in 16 patients were compared point-by-point with MRI data (linear regression and Bland-Altman analyses). Global LVV correlated highly with MRI ( r =0.98; y =0.99 x +2.3) with minimal bias (1.4 mL) and narrow limits of agreement (±20 mL). WM correlated highly only in basal and midventricular segments ( r =0.88; y =0.85 x +0.7). In protocol 2, we tested the ability of this technique to differentiate populations with known differences in LV function by studying 9 patients with dilated cardiomyopathy and 9 normal subjects. All calculated indices of global and regional systolic and diastolic LV function were significantly different between the groups. In protocol 3, we tested the feasibility of automated detection of regional WM abnormalities in 11 patients. In each segment, abnormality was detected when regional shortening fraction was below a threshold obtained in normal subjects. The automated detection agreed with expert interpretation of 2D WM in 86% of segments.

Conclusions— Volumetric analysis of RT3DE data is clinically feasible and allows fast, semiautomated, dynamic measurement of LVV and automated detection of regional WM abnormalities.

Automated quantitative assessment of wall motion in patients with poor acoustic windows

BACKGROUND

No technique exists for objective evaluation of left ventricular wall motion in contrast-enhanced images. We tested a new technique for quantification of regional fractional area change using contrast-enhanced power modulation imaging with color kinesis.

METHODS

The feasibility of this technique for detecting acute ischemia was first tested in 11 pigs. Next, the accuracy for detecting resting wall-motion abnormalities was determined in 52 patients requiring contrast and compared with conventional interpretation of 2-dimensional images by inexperienced readers. Expert interpretation of 2-dimensional images served as the gold standard.

RESULTS

In pigs, coronary occlusion resulted in reversible hypokinesis and reduced regional fractional area change. In patients with poor acoustic windows, this technique's accuracy for quantitative detection of resting wall-motion abnormalities was 86% compared with 81% for conventional interpretation by inexperienced readers (P <.01).

CONCLUSIONS

Regional wall motion can be accurately assessed using color-encoded power modulation imaging for patients requiring contrast. This technique may prove a useful diagnostic aid to echocardiographers of varying levels of experience.

Quantitative assessment of right ventricular systolic function by the analysis of right ventricular contrast time-intensity curve

To study reliability and reliable indices of quantitative assessment of right ventricular systolic function by time-intensity curve (TIC) with right ventricular contrast, 5% sonicated human albumin was injected intravenously at a does of 0.08 ml/kg into 10 dogs at baseline status and cardiac insufficiency. Apical four-chamber view was observed for washin and washout of contrast agent from right ventricle. The parameters of TIC were obtained by curve fitting. The differences of parameters were analyzed in different states of cardiac functions. Among the parameters derived from TIC, the time constant (k) was decreased significantly with decline of cardiac function (P<0.001). But half-time of decent of peak intensity (HT) and mean-transit-time (MTT) of washout were increased significantly (P<0.001). The k was strongly related to cardiac output of right ventricle (CO) and ejection fraction (EF) of left ventricle and fractional shortening (FS) of left ventricle. Right ventricular systolic function could be assessed reliably by the parameters derived from TIC with right ventricular contrast echocardiography. The k, HT and MTT are reliable indices for quantitative assessment of right ventricular systolic function.

Interpretation of Left Ventricular Wall Motion During Stress Testing

This article describes the obstacles to stress echocardiographic interpretation, and reviews the techniques currently available that offer a more objective approach to stress wall motion analysis than the conventional visual methodology. These techniques include Doppler-based methods, such as myocardial Doppler velocity and strain rate imaging, as well as automated border detection techniques, such as acoustic quantification and color kinesis.

Automated quantification of left ventricular function by the automated contour tracking method

The automated contour tracking (ACT) method has been developed for the automated measurement of area volume using the energy minimization method without tracing a region of interest. The purpose of this study was to compare the ACT method and left ventriculography (LVG) for the measurement of left ventricular (LV) function in the clinical setting. An apical four-chamber view was visualized by two-dimensional echocardiography and recorded for off-line analysis in 14 patients with high-quality images who underwent LVG. The ACT method automatically traces the endocardial border from the recorded images and calculates LV volumes (end-diastole and end-systole) and ejection fraction (EF). Both ACT and LVG were compared by linear regression analysis for the measurement of EF. EF determined by the ACT method agreed well with that by LVG (r = 0.96, y = 0.94x + 4.6, standard error of the estimate = 3.9%). The mean difference between the ACT and LVG was -1.4%+/- 7.3%. In conclusion, the ACT method is reliable for noninvasive estimation of EF in high-quality images. This suggests that this new technique may be useful in the automated quantification of LV function.

Analysis of cardiac left-ventricular volume based on time warping averaging

The cardiac left-ventricular (LV) volume signal, obtained by acoustic quantification, is affected by noise and respiratory modulation, resulting in a large beat-to-beat variability that affects the computation of LV function indices. A new method is proposed to improve the evaluation of LV indices by applying a signal averaging technique based on dynamic time warping to consecutive LV volume waveforms. Volume signals obtained from ten normal young (NY) subjects (mean age +/- SD: 25+/-5 years) were used to evaluate the performance of this algorithm. To evaluate its clinical utility, the effects of ageing and pharmacologically induced changes on LV function were assessed by studying, respectively, ten normal (N) adult subjects (age 64+/-8 years) and ten patients with dilated cardiomyopathy during a control and low-dose dobutamine (10 microg kg(-1) min(-1)) study. Indices of LV function were highly consistent, with a variability of less than 8%, even when only 16 beats were averaged, independently of their selection inside the whole recording. When compared with beat-to-beat measures, the averaging of 16 beats significantly reduced (by more than 50%) the interbeat variability of all indexes. Expected alterations in both diastolic and systolic function were evidenced both with ageing (peak filling atrial contraction and ejection rates: from 275+/-77 mls(-1), 76+/-30 ml s(-1) 230+/-70 mls(-1), respectively, in NY, to 160+/-33 mls(-1), 125+/-39 mls(-1), 163+/-54 mls(-1) in N) and with dobutamine (peak filling and ejection rates from 160+/-72 mls(-1) and 183+/-86 mls(-1) respectively, in control, to 253+/-75 mls(-1) and 251+/-105 mls(-1) with dobutamine). Signal averaging with time warping allows fast and improved assessment of LV function.

Normal values of regional left ventricular endocardial motion: multicenter color kinesis study

Our goal was to establish normal values for quantitative color kinesis indexes of left ventricular (LV) wall motion over a wide range of ages, which are required for objective diagnosis of regional systolic and diastolic dysfunction. Color-encoded images were obtained in 194 normal subjects (95 males, 99 females, age 2 mo to 79 yr) in four standard views. Quantitative indexes of magnitude and timing of systolic and diastolic function were studied for age- and gender-related differences. Normal limits of all ejection and filling indexes were in a narrow range (< or =25% of the mean), with no major gender-related differences. Despite invariable ejection fractions, both peak filling and ejection rates decreased with age (30 and 20%, correspondingly) with a concomitant increase in mean filling and ejection times, resulting in five- and twofold increases in the late to early filling and ejection ratios, correspondingly. Diastolic asynchrony increased with age (from 4.7 +/- 2.0 to 6.4 +/- 3.2 from the 2nd to 7th decade). The normal values of color kinesis indexes should allow objective detection of regional LV systolic and diastolic dysfunction.

Assessment of elastic properties of the descending thoracic aorta by transesophageal echocardiography with acoustic quantification in patients with a stroke

Previous studies have described the use of transesophageal echocardiography (TEE) with acoustic quantification (AQ) in assessing aortic elastic properties. We hypothesized that patients with a prior history of stroke (ST) may have a higher risk of atherosclerotic change in great vessels compared to nonstroke subjects (NST) and thus have decreased elastic properties. We assessed the elastic properties of the descending thoracic aorta (DTA) by TEE in ST patients and compared them with data in NST patients. Subjects included 31 with ST without any evidence of emboli originating from the heart (age 51 +/- 10 years, M:F = 20:11) and 25 age-matched NST (M:F = 8:17). Patients with significant valvular heart disease including aortic and mitral regurgitation, left ventricular dysfunction (ejection fraction < 55%), and congenital heart disease were excluded. Compliance (C), distensibility (D), and stiffness index (SI) were measured using AQ and M-mode measurement at a level of the left atrium. We scored atherosclerotic risk factors (ARF) such as a history of diabetes, hypertension, smoking, hypercholesterolemia, and the presence of atheroma of DTA. There was no evidence of atheroma of DTA in NST. There were no significant differences in heart rate and systolic and diastolic blood pressure between ST and NST patients. Fractional area change (FAC) of DTA was significantly lower in ST than in NST patients (3.2 +/- 1.6 vs 5.4 +/- 2.5%, P = 0.000). ST patients had significantly lower C (1.2 +/- 0.4 vs 1.5 +/- 0.7 x 10(-3) cm2 mmHg(-1), P = 0.039), lower D (0.8 +/- 0.3 vs 1.5 +/- 0.8 x 10(-3) mmHg(-1), P = 0.000), and higher SI (10.3 +/- 8.8 vs 5.3 +/- 2.9, P = 0.006) than NST patients. ST patients without atheroma of DTA (n = 21) also had significantly lower C (1.1 +/- 0.4 vs 1.5 +/- 0.7 x 10(-3) cm2 mmHg(-1), P = 0.038) and lower D (3.5 +/- 1.4 vs 4.8 +/- 2.4 x 10(-3) mmHg(-1), P = 0.021) than NST patients. There was a significant positive correlation between SI and the score of ARF (r = 0.51, P = 0.000). The regional elastic properties of DTA measured by TEE with AQ and M-mode method were abnormal in ST. Therefore, TEE with AQ technique may have a possible clinical application for the detection of early atherosclerotic changes such as alteration of elastic properties in morphological normal DTA.

Left ventricular pressure-area relations as assessed by transoesophageal echocardiographic automated border detection: comparison with conductance catheter technique in cardiac surgical patients

The aim of this study was to validate measurements of intraoperative left ventricular (LV) area by transoesophageal echocardiography against simultaneous measurements of LV volume by conductance catheter (CC) in cardiac surgical patients with normal systolic LV function. Echo area was compared with CC volume during steady state and during acute changes of pre- and afterload by partial clamping of the inferior vena cava and the ascending aorta in eight patients scheduled for coronary artery bypass grafting. At steady state, Bland-Altman analysis of 32 recordings revealed a bias (SD) of 0.6% (2.5%) between echo area and CC volume, related to the initial values of end-diastolic area (100% area) and volume (100% volume), respectively. During loading interventions, bias between the two methods, as assessed by 112 measurement sequences, was 0.5% (3.7%) during aortic occlusion and -3.9% (4.4%) during cava occlusion at end-systole (P < 0.0001); at end-diastole, this bias was 1.3% (4%) during aortic occlusion and 0.2% (5.7%) during cava occlusion (P < 0.0001). Intraoperative area measurements with transoesophageal echocardiography in cardiac surgical patients with normal systolic LV function show good correlation with CC volume measurements under steady-state conditions. During acute unloading by vena cava occlusion, the resulting small end-systolic echo area measurements differ significantly more from CC volume measurements than during acute increase in afterload by aortic occlusion.

Validity and reproducibility of echocardiographic measurement of left ventricular ejection fraction by acoustic quantification with tissue harmonic imaging technique

The tissue harmonic imaging technique can enhance detection of the cardiac endocardial border. When combined with an acoustic quantification (AQ) method, an improvement of accuracy and reproducibility of real-time measurement of left ventricular (LV) function might be expected. However, few data exist regarding the measurement of LV function by AQ with the harmonic imaging technique. Therefore, we evaluated the validity and reproducibility of AQ measurement of LV ejection fraction with or without harmonic imaging technique. A total of 50 patients (mean age 58 +/- 10 years) who underwent left ventriculography were included in our study. The LV end-diastolic and end-systolic volumes by ventriculography were 131 +/- 52 mL and 72 +/- 43 mL, respectively, and were underestimated by both conventional (70 +/- 32 mL and 36 +/- 25 mL) and harmonic (67 +/- 30 mL and 34 +/- 22 mL) AQ obtained in the apical 4-chamber view. The calculated ejection fraction by ventriculography was 0.49 +/- 0. 11 and correlated with that by conventional AQ (0.51 +/- 0.11; y = 0. 72x + 0.152; r = 0.73). This was a marked improvement when compared with the ejection fraction by harmonic AQ (0.50 +/- 0.11; y = 0.89x + 0.065; r = 0.91). Interestingly, interobserver and intraobserver variabilities of conventional AQ, which were 15.6% and 8.6%, respectively, were much improved by harmonic AQ (8.9% and 4.5%, respectively). These results indicate the feasibility of real-time measurement of LV ejection fraction by harmonic imaging, although absolute LV volume can be underestimated even by this technique.

Evaluation of respiratory influences on left ventricular function parameters extracted from echocardiographic acoustic quantification

This study was designed to assess, using the echocardiographic acoustic quantification technique, the influence of respiration on left ventricular (LV) function and its modifications connected with the ageing process, quantifying in a non-invasive way the respiratory contribution to the LV volume variability. An automated algorithm is applied to extract the beat-to-beat measurements of LV function parameters from the LV volume signal, obtained from recordings lasting a few minutes. Mean values, amount of variability and spectral content were studied in a population of 17 normal young (mean age 25 +/- 1 years) and 12 normal old (mean age 64 +/- 2 years) subjects. Mean values of the beat-to-beat measurements of LV function parameters were able to point out alterations connected with the ageing process in peak filling rate, peak atrial filling rate and peak ejection rate. Spectral analysis, applied to the extracted variability series, displayed a predominance of the high-frequency (HF) component corresponding to respiration in all LV function parameters; moreover, age related changes of HF variability were observed in peak ejection rate. The HF power spectrum component of beat to beat series extracted from the LV signal can provide a non-invasive assessment of the fluctuations in ventricular parameters associated with respiration.

Impact of on-line endocardial border detection on determination of left ventricular volume and ejection fraction by transthoracic 3-dimensional echocardiography

This study was performed to determine whether use of on-line automated border detection (ABD) could reduce data analysis time for 3-dimensional echocardiography (3DE) while maintaining accuracy of 3DE in measures of left ventricular (LV) volumes and ejection fraction (EF). The study proceeded in 2 phases. In the validation phase, 20 subjects were examined with the use of 3DE and of monoplane 2-dimensional (2D) ABD. Results were compared with the reference standard of magnetic resonance imaging (MRI). In the test phase, 20 subjects underwent two 3DE studies (once with images optimized for visual border definition and once with images optimized for ABD border tracking) and a conventionally used 2D ABD study. For 3DE, volumes and EF were determined with the use of manually traced borders and ABD. Analysis times were recorded with a digital stopwatch. In the validation phase, 3DE and MRI results correlated very well (r = 0.99) without systematic differences. Comparison of 2D ABD with MRI showed good correlation for LV volumes (r >/= 0.90) and EF (r = 0.85) despite significant underestimation. For the test phase, Acoustic Quantification-optimized 3-dimensional datasets underestimated end-diastolic volume and EF relative to visually optimized 3-dimensional datasets regardless of whether borders were hand-traced or ABD was used. However, correlations ranged from r = 0.96 to r = 0.98 for LV volumes and 0.88 to 0.91 for LV EF and were superior to those for 2D ABD. Data analysis times decreased moderately with the use of ABD, but scan times increased; total study times were unchanged. Use of on-line ABD with 3DE reduces data analysis time and is more accurate than conventional monoplane 2D ABD but results in underestimation of LV volumes and EF. Additional automated postprocessing techniques may be required to obtain accurate measures, consistently using 3DE in conjunction with on-line ABD.

Evaluation of the Left Atrial Performance Using Acoustic Quantification

In most clinical studies, evaluation of left atrial (LA) function evolved from estimation of LA size on chest radiograph, electrocardiographic P wave abnormalities, LA diameter determined at fluoroscopy or by echocardiography, LA pressure recordings, blood flow velocity with Doppler echocardiography, and measurements of LA volume based on echocardiographic, cineangiographic, radionuclide, and magnetic resonance imaging techniques. The recent development of real-time two-dimensional echocardiographic acoustic quantification (AQ) suggests that LA dimensions can be measured instantaneously to provide online assessment of its systolic and diastolic functions. By means of AQ echocardiography and simultaneous recordings of LA pressure, the LA pressure-area relation can be obtained. LA pressure-area relation consisted of two loops: the A loop, representing the LA pump function, and the V loop, representing LA reservoir and passive emptying functions. The importance of LA function has been demonstrated in congestive heart failure, atrial fibrillation, hypertension, and ischemic heart disease and during pacing or dobutamine infusion.

Online Quantification of Left Ventricular Function: Correlation with Various Imaging Modalities

The need for more objectively quantifiable evaluation of the left ventricular function, obtainable on line with echocardiography, was fulfilled through modifications of integrated backscatter imaging to permit real-time differentiation of the endocardium and the blood pool area in every image frame. Operator-guided study of individual cardiac chambers permitted instantaneous measurement of chamber areas in either the systole or diastole, with resulting physiologically meaningful recordings that relate to cardiac function. Validation studies by various approaches suggest that the methodology is clinically relevant, and further improvements in design will sharpen its applicability in the future.

Acoustic Quantification Today and Its Future Horizons

We briefly review previously published work based on the uses of acoustic quantification (AQ) or validation of this technology. We also discuss the limitations of AQ in a critical review of the literature, including operator dependency, signal noise, and low temporal resolution. We describe some enhancements made to AQ software to address these limitations and improve the accuracy of this technique, including digital beam processing, harmonic imaging, and signal averaging. Several anticipated applications are also briefly described for those interested in the future development of this technology. These future applications include noninvasive long-term monitoring of ventricular function and objective assessment of regional ventricular wall motion in two and three dimensions.

Load-Independent Indices of Left Ventricular Function Using Automated Border Detection

Echocardiographic automated border detection is the ability to assess left ventricular (LV) cross-sectional cavity areas and volumes on line. This capability has enabled the use of LV pressure-volume relationships as a means to determine LV function in a manner relatively independent of loading conditions. This discussion reviews previous validation studies in animal models and humans using LV cross-sectional area as a surrogate for LV volume and the clinical applications of LV pressure-area relations. Applications of arterial pressure as a substitute for LV ejection pressure to assess pressure-area relations are also reviewed, along with the use of pressure-area relations to assess right ventricular performance. Last, preload-adjusted maximal power as an alternative load-insensitive means to determine LV performance is discussed.

Estimation of Left Ventricular Ejection Fraction by Semiautomated Edge Detection

Left ventricular (LV) volume and ejection fraction estimation from two-dimensional echocardiograms requires off-line analysis and time-consuming manual tracing. LV volumes may be estimated on-line with a semiautomated edge detection echocardiographic system [also known as acoustic quantification (AQ)], but there are few data that compare volumes obtained from the AQ method with volumes derived from off-line manual tracing of conventional two-dimensional echocardiograms. Echocardiograms were performed in 48 patients at two medical centers. LV volumes were measured from the apical view with the method of discs and area-length formulae and from the parasternal short-axis view with the modified ellipsoid model. Based on the criterion of >/=75% endocardial visualization, 25 (52%) of the short-axis views and 14 (29%) of the apical views were analyzed by a single investigator. End-diastolic and end-systolic LV volumes derived on line with the AQ system showed a very strong linear association with off-line, manually traced volumes (r = 0.96-0.99). Correlations for ejection fraction also were strong (r = 0.90-0.96). End-diastolic and end-systolic LV volumes, measured from the apical views, were underestimated by the AQ method. However, because the error was in the same direction, ejection fractions measured with the AQ system and by manual tracing of conventional echocardiograms were similar. Estimation of ejection fraction using a semiautomated edge detection echocardiographic system is a promising method for noninvasive evaluation of systolic function in carefully selected patients.

Comparison of automatic boundary detection and manual tracing technique in echocardiographic determination of left atrial volume

Previous reports have indicated that echocardiography with automatic boundary detection (ABD) is useful for the noninvasive estimation of left ventricular volume. However, few data exist regarding the measurement of left atrial (LA) volume, which also provides pivotal information in the clinical setting. Therefore, the feasibility of LA volume measurement by ABD in comparison with the manual tracing using modified Simpson's method (SM) was evaluated. Fifty-nine patients with coronary artery-disease with sinus rhythm were examined. Using ABD, a region of interest was set around the LA border and mitral annulus from an apical four-chamber view. The maximal and minimal LA volume (Vmax and Vmin) were measured from the volume waveform. Using the SM, the maximal and minimal LA volume were measured by the manual tracing on frozen frames at the apical four-chamber view. The ABD displayed a curve of LA volume change that consisted of passive emptying, diastasis, and active emptying phases during the left ventricular diastolic period. Under these conditions, the Vmax and Vmin were 43.7 +/- 11.2 ml and 21.1 +/- 7.6 ml, respectively, yielding the volume change of 22.6 +/- 6.0 ml. By the SM, Vmax and Vmin were 43.1 +/- 9.9 ml (r = 0.94, p < 0.0001, y(ABD) = 0.91x (SM) + 3.6) and 22.0 +/- 9.0 ml (r = 0.91, p < 0.0001, y = 0.94x + 0.7), respectively, and the volume change was 22.8 +/- 6.1 ml (r = 0.82, p < 0.0001, y = 0.84x + 3.8). These results indicate that the ABD from the apical four-chamber approach could provide an accurate estimation of LA volume change, suggesting the potential value of this method in assessing LA function, although some technical difficulties need to be further overcome.

Acoustic quantification indexes of left ventricular size and function: effects of signal averaging

BACKGROUND

The aim of this study was to evaluate the clinical utility of using signal-averaged acoustic quantification (SAAQ) waveforms for improved assessment of left ventricular (LV) size and function.

METHODS AND RESULTS

Four separate protocols were performed in 47 subjects. SAAQ waveforms were used to assess alterations in LV function induced by dobutamine (15 microg/kg per minute) and esmolol (200 microg/kg per minute) in eight normal subjects. Subsequently, we analyzed SAAQ waveforms obtained in 12 patients with LV dysfunction secondary to dilated cardiomyopathy and 12 age-matched normal subjects. Finally, we developed computer software for monitoring of LV function on the basis of continuous acquisition and repeated analysis of SAAQ waveforms. We compared the interbeat variability in indexes of LV function obtained from raw AQ and SAAQ during 10 minutes of steady-state monitoring in eight patients undergoing transesophageal echocardiography. The feasibility of long-term monitoring in the intensive care setting was then studied in seven patients undergoing abdominal surgery. Our analysis tracked variations in LV function induced by dobutamine and esmolol. Significant differences in all measured indexes were found between normal subjects and patients with dilated cadiomyopathy. Signal averaging during steady-state monitoring significantly reduced the interbeat variability of all indexes (21% to 42%). In the operating room, the SAAQ monitoring system tracked hemodynamic changes in close agreement with invasive measurements.

CONCLUSIONS

SAAQ allows fast and easy quantification of LV function and can track hemodynamic trends in the operating room setting.

Critical appraisal of left ventricular function assessment by the automated border detection method on echocardiography. Is it good enough?

Many studies have attempted to validate the echocardiographic automated border detection (ABD) method for assessing left ventricular ejection fraction (LVEF) by comparing it with various echocardiographic and non-echocardiographic standards. The main basis of assessing its accuracy has been the coefficient of correlation. The fallacy of using coefficient of correlation for assessing agreement between two methods of measurement has been well emphasized in the literature. In the present study we used the Bland and Altman test for testing the accuracy of the ABD method. We compared the ABD method for LVEF assessment with the manual edge detection technique on echocardiography and with radionuclide ventriculography in 34 patients. The majority of patients (76%) had regional wall motion abnormality. The ABD method could be adequately performed in 25 (74%) patients. LVEF was significantly underestimated by the ABD method with very wide limits of agreement when compared with radionuclide ventriculography and the manual edge detection technique (-9.2+/-21.7 and -2.7+/-18.4 respectively, mean error+/-2 standard deviations). Stated simply, the ABD method could overestimate LVEF by 12.5 and 15.7 or underestimate by 30.9 and 21.1 when compared with radionuclide ventriculography and manual edge detection technique, respectively. This large error is by no means acceptable for clinical purposes. It is concluded that at the present stage, the ABD method cannot replace radionuclide ventriculography and manual edge detection technique for assessing LVEF.

New modalities of regional and global left ventricular function analysis: state of the art

Study of left ventricular (LV) regional and global function represents a main point of the cardiologic evaluation. This article presents an overview of the state of the art in quantitative analysis of ventricular wall motion and describes the different techniques available in clinical settings; we also present the personal experience of the authors in comparing conventional 2-dimensional (2D) echocardiography with other recently developed, more sophisticated techniques. Conventional 2D echocardiography mainly depends on the operator's ability. Moreover, the physiologic regional pattern of myocardial wall motion in different segments of left ventricle is still not completely known as well as the heart's rotational and translational movements. Qualitative and quantitative transesophageal echocardiography allows a better and more accurate evaluation of regional wall motion, and improves the on-line border detection feasibility even in patients with poor transthoracic echocardiographic window. The automated system for on-line endocardial border detection, color kinesis, and the power Doppler are, at the moment, promising techniques. Using magnetic resonance imaging as a "gold standard" for the study of global and regional left ventricular function, the investigators describe personal experiences with tissue Doppler imaging and a new computerized system for tissue Doppler images postprocessing analysis.

Color Kinesis: Principles of Operation and Technical Guidelines

Color kinesis is a new echocardiographic technique that aids in the assessment of global and regional left ventricular performance during either systole or diastole. Color kinesis uses automated border detection technology based on backscatter data to display both the magnitude and timing of endocardial motion in real time. The color kinesis display superimposes a color overlay on the two-dimensional echocardiographic image; the number of color pixels represents the magnitude of endocardial motion, while the different colors represent the timing of endocardial motion according to a predefined color scheme. Because color kinesis is an operator-dependent technique, the steps involved in performing a technically adequate study will be reviewed as well as the pitfalls and technical limitations. The potential clinical applications of color kinesis will also be discussed.

Noninvasive markers for acute heart transplant rejection in children with the use of automatic border detection

A noninvasive method to detect heart transplant rejection would allow for increased monitoring at reduced risk. Automatic border detection is a new method to assess diastolic abnormalities. The purpose of this study was to determine whether automatic border detection of left ventricular filling detects rejection in children. Nineteen episodes of biopsy-proven rejection in 10 children were retrospectively reviewed. Echocardiograms during rejection were compared with those before rejection and during recovery. Automatic border detection indices were percentage of total left ventricular filling as a result of rapid filling, diastasis, and atrial contraction. The percentage of total ventricular filling during diastasis increased significantly during rejection (10% +/- 6% versus 15% +/- 8%, p = 0.02), and the percentage of filling during the rapid filling phase decreased during rejection (82% +/- 8% versus 77% +/- 11%, p = 0.08). These changes were even more marked for the most severe episodes of rejection. These changes resolved at recovery. Automatic border detection of left ventricular filling patterns are altered during cardiac rejection in children. Filling during diastasis increases significantly, and filling during the rapid filling phase decreases. A prospective analysis is needed to determine whether these changes in filling can obviate the need for cardiac biopsy.

Comparison of continuous left ventricular volumes by transthoracic two-dimensional digital echo quantification with simultaneous conductance catheter measurements in patients with cardiac diseases

Automated border detection enables real-time tracking of left ventricular (LV) volume by 2-dimensional transthoracic echocardiography. This technique has not been previously compared with simultaneously measured continuous LV volumes at rest or during transients in humans. We performed 18 studies in 16 patients (age 50 +/- 15 years, range 22 to 70; ejection fraction 63 +/- 20%, range 15% to 85%) in which continuous LV volumes acquired by digital echo quantification (DEQ) were compared with simultaneous conductance catheter volume obtained by cardiac catheterization. Both volume signals were calibrated by thermodilution-derived cardiac output and ventriculogram-derived ejection fraction. Volume traces acquired at rest were averaged to generate a comparison cycle. The averaged volume waveforms acquired by DEQ and by conductance catheter were similar during all phases of the cardiac cycle and significantly correlated (conductance catheter = slope. DEQ + intercept, slope = 0.94 +/- 0.09, intercept = 5 +/- 8 ml, r2 = 0.86 +/- 0.12, all p <0.0001). Steady-state hemodynamic parameters calculated using either averaged volume signal were significantly correlated. Transient obstruction of the inferior vena cava yielded a 45 +/- 13% decrease in end-diastolic volume. Successful recordings of DEQ volume during preload reduction were obtained in only 50% of studies. End-diastolic volumes from the 2 methods were significantly correlated (mean slope 0.88 +/- 0.31, mean intercept 14 +/- 37 ml, average r2 = 0.89 +/- 0.11, all p <0.01), as were end-systolic volumes: mean slope 0.80 +/- 0.43, intercept = -20 +/- 26 ml, r2 = 0.67 +/- 0.18, all p <0.05). We conclude that automated border detection technique by DEQ is reliable for noninvasive, transthoracic, continuous tracking of LV volumes at steady state, but has limitations in use during preload reduction maneuvers in humans.

Freehand three-dimensional echocardiography for measurement of left ventricular mass: in vivo anatomic validation using explanted human hearts

OBJECTIVES

We sought to validate freehand three-dimensional echocardiography for measuring left ventricular mass and to compare its accuracy and variability with those of conventional echocardiographic methods.

BACKGROUND

Accurate measurement of left ventricular mass is clinically important as a predictor of morbidity and mortality. Freehand three-dimensional echocardiography eliminates geometric assumptions used by conventional methods, minimizes image positioning errors using a line of intersection display and increases sampling of the ventricle. Preliminary studies have shown it to have high accuracy and low variability.

METHODS

Twenty-eight patients awaiting heart transplantation were examined by conventional and freehand three-dimensional echocardiography. Left ventricular mass was determined by the M-mode ("Penn-cube") method, the two-dimensional truncated ellipsoid method and three-dimensional surface reconstruction. The ventricles of 20 explanted hearts were obtained, trimmed and weighed. Echocardiographic mass by each method was compared with true mass by linear regression. Accuracy, bias and interobserver variability were calculated.

RESULTS

For three-dimensional echocardiography, the correlation coefficient, standard error of the estimate, root mean square percent error (accuracy), bias and interobserver variability were 0.992, 11.9 g, 4.8%, -4.9 g and 11.5%, respectively. For the two-dimensional truncated ellipsoid method they were 0.905, 38.5 g, 15.6%, 15.4 g and 23.3%. For the M-mode ("Penn-cube") method they were 0.721, 96.9 g, 53.0%, 109.2 g and 19.5%.

CONCLUSIONS

Freehand three-dimensional echocardiography for measurement of left ventricular mass has high accuracy and low variability and is superior to conventional methods in hearts of abnormal size and geometry.

Detection of diastolic dysfunction: acoustic quantification (AQ) in comparison to Doppler echocardiography

OBJECTIVES

To evaluate the potential of acoustic quantification (AQ) in detection of diastolic dysfunction in comparison to Doppler analysis, we investigated, as a model of restrictive filling pattern, nonrejecting heart transplant recipients early postoperatively.

BACKGROUND

AQ, an ultrasonic backscatter imaging system, enables instantaneous calculation of cavity areas and thus provides a new approach to diastolic function.

METHODS

Of 27 pts who have undergone heart transplantation, echocardiography has been performed at the day of biopsy. During a time course of 8 weeks echocardiographic data have been analysed at 3 different time points (early, mid and late) in 16 nonrejecting pts. Indexes of the area-change waveform and its 1. derivative (dA/dt) obtained by AQ were opposed to usual Doppler indexes.

RESULTS

In comparing data of the early and late time point of investigation, significant changes of early diastolic filling were detectable by AQ as well as by Doppler: End-diastolic areas have increased (p < 0.001), while peak filling rate (p < 0.0001), slope of area change during rapid filling (p < 0.001) and amount of relative area change during rapid filling (p < 0.001) have decreased. Complementary, Doppler derived pressure half-time (p < 0.0001) and isovolumic relaxation time (p < 0.0001) have increased while the peak early filling velocity (p < 0.0001) and its time velocity integral (p < 0.001) have decreased.

CONCLUSION

An initial restrictive filling pattern has improved 8 weeks postoperatively. Since multiple indexes, obtained from the area change waveforms, in particular the for end-diastolic area normalized peak filling rate, seem to be highly sensitive in detecting changes of diastolic function, AQ may play an important complementary role in non-invasive evaluation of restrictive filling pattern.

Right ventricular volumes overestimate left ventricular preload in critically ill patients

BACKGROUND

Studies have shown right ventricular end-diastolic volume (RVEDV) to be a more accurate estimate of left ventricular preload than pulmonary artery wedge pressure. We prospectively evaluated the ability of RVEDV to predict left ventricular end-diastolic volume (LVEDV) in critically ill patients.

METHODS

Thirty critically ill patients in the surgical intensive care unit underwent concurrent measurement of RVEDV and LVEDV. RVEDV was measured using a residual fraction Swan-Ganz catheter (RF Swan). LVEDV was measured using transesophageal echocardiography with acoustic quantification. Intracardiac, intra-abdominal, and ventilatory pressures were also measured.

RESULTS

RVEDV as measured by the RF Swan was significantly larger (by a factor of 2) than LVEDV (p < 0.0001 analysis of variance). However, the RVEDV and LVEDV were strongly correlated (r = 0.71, p < 0.0001, Pearson's correlation).

CONCLUSIONS

RVEDV from the RF Swan markedly overestimated left ventricular preload. If RVEDV is used as an absolute value for determining preload, patients may be underresuscitated. Transesophageal echocardiography in conjunction with RF Swan can be used to more accurately determine preload and cardiac performance than RF Swan alone in critically ill patients.

Two-dimensional Fourier filtration of acoustic quantification echocardiographic images: improved reproducibility and accuracy of automated measurements of left ventricular performance

To determine the accuracy of Fourier filtration in removing the high-frequency component of noise from acoustic quantification (AQ) echocardiographic images, we processed 800 parasternal short-axis images obtained from 10 study subjects. M-mode tracings were also obtained and used as gold standard for correlating the results from raw AQ and Fourier-filtered images. Left ventricular short-axis diameters measured from the raw AQ and Fourier-filtered data were compared with the M-mode diameters (r = 0.91, p < 0.001 for raw AQ; and r = 0.96, p < 0.001, for Fourier filtered images). Fractional shortening showed better correlation between Fourier-filtered images and M-mode (r = 0.79, p < 0.03) versus raw AQ and M-mode (r = 0.33, p = 0.46). Best-to-beat reproducibility was also found to be better for fractional area change (r = 0.82, p = 0.01 versus r = 0.66, p = 0.77), peak area filling rate (r = 0.87, p = 0.004 versus r = 0.62, p = 0.1), and peak are emptying rate (r = 0.99, p < 0.0001 versus r = 0.19, p < 0.7) for Fourier filtered versus raw AQ. Our results indicate that Fourier filtration of AQ data results in more accurate representation of the true endocardial borders.

Segmental analysis of color kinesis images: new method for quantification of the magnitude and timing of endocardial motion during left ventricular systole and diastole

BACKGROUND

We describe a method for objective assessment of left ventricular (LV) endocardial wall motion based on Color Kinesis, a new echocardiographic technique that color-encodes pixel transitions between blood and myocardial tissue.

METHODS AND RESULTS

We developed a software that analyzes Color Kinesis images and provides quantitative indices of magnitude and timing of regional endocardial motion. Images obtained in 12 normal subjects were used to evaluate the variability in each index. Esmolol, dobutamine, and atropine were used to track variations in LV function in 14 subjects. Objective evaluation of wall motion was tested in 20 patients undergoing dobutamine stress testing. Regional fractional area change, displacement, and radial shortening were displayed as histograms and time curves. Global function was assessed by calculating magnitude and timing of peak ejection or filling rates and mean time of ejection or filling. Patterns of endocardial motion were consistent between normal subjects. Fractional area change and peak ejection rate decreased with esmolol and increased with dobutamine. Time to peak ejection and mean time of contraction were prolonged with esmolol and shortened with dobutamine. Using atropine, we proved that our findings with dobutamine were not secondary to its chronotropic effects. Dobutamine induced regional wall motion abnormalities in 10 patients in 38 segments diagnosed conventionally. Segmental analysis detected abnormalities in 36 of these 38 segments and in an additional 5 of 322 segments.

CONCLUSIONS

Analysis of Color Kinesis images allows fast, objective, and automated evaluation of regional wall motion sensitively enough to evaluate clinical dobutamine stress data. This method has significant potential in the diagnosis of myocardial ischemia.

Detection of cardiac boundaries in echocardiographic images using a customized order statistics filter

Order statistics filters are nonlinear filters that suppress impulsive and Gaussian noise while preserving edges. These features are particularly useful for cardiac boundary detection in ultrasound images. Based on these facts, we have analyzed performance of a combined ranked order statistics filter. The filter subtracts ranks of ordered highest and lowest intensity values of pixels encompassed in a filter window. The rank extent and window size selection allow adjustment of filter properties for a particular application. Increasing the rank fosters the low-pass characteristics of the filter. Increasing the window size supports noise removal but reduces anatomic selectivity. The filter highlights cardiac boundaries in clinical echocardiograms with intensity proportional to the local probability of a presence of the boundary.

Evaluation of color kinesis, a new echocardiographic method for analyzing regional wall motion in patients with dilated left ventricles

The recently developed echocardiographic technology of color kinesis (CK) displays endocardial motion in color layers on a single end-systolic 2-dimensional echocardiographic frame. Previous work using this method is promising for quantitation of regional function, but there is limited experience in patients with severely reduced left ventricular function. Twenty patients (age 59 +/- 10 years) with dilated cardiomyopathy (left ventricular ejection fraction 22 +/- 8%) underwent CK imaging. Endocardial motion was quantitated by measuring the distance of endocardial motion during the systolic interval and also by calculating the endocardial velocity. CK measurements were compared among 4 wall motion grades (i.e., normal, hypokinetic, akinetic, and dyskinetic) assessed by qualitative wall motion scoring. There was a significant overall difference (p < 0.0001) in the mean systolic endocardial inward motion (i.e., contraction) and outward motion (i.e., expansion) among wall motion grades. The mean endocardial outward distance was significantly greater for the dyskinetic segments than for the other grades (p < 0.001). There were also differences in the mean velocity of endocardial motion among the wall motion grades. In the presence of left bundle branch block, there was no difference in the mean endocardial inward distance of the hypokinetic, akinetic, and dyskinetic septal segments. We conclude that in the absence of left bundle branch block, normal, hypokinetic, akinetic, and dyskinetic ventricular wall segments may be distinguished in patients with dilated cardiomyopathy on the basis of endocardial motion measured with CK.

Comparison of echocardiographic acoustic quantification system and radionuclide ventriculography for estimating left ventricular ejection fraction: validation in patients without regional wall motion abnormalities

Echocardiographic automated border detection of blood-endocardium interface is made on the basis of the principle of acoustic quantification. The automated border system is capable of providing on-line left ventricular (LV) cavity area and function. Recently, ABD algorithms have been devised to estimate LV volume on line from a long-axis image, calculated by established area-length method or Simpson's formula. To test the clinical validity of this newly developed echocardiographic method, LV volumes and ejection fraction measured by real-time acoustic quantification were compared with radionuclide ejection fraction in 24 subjects on the same day. Patients were included in the study if > or = 75% of their endocardium was visualized with conventional two-dimensional echocardiography. Sixteen (66%) of 24 patients had a technically adequate conventional echocardiogram with a broad range of ventricular dimensions and systolic function. None of the study patients had regional wall motion abnormalities. Echocardiographic measurements were obtained from the LV apical four-chamber, long-axis view. Ejection fraction, determined by the acoustic quantification and by radionuclide ventriculography, showed a strong linear relation (r = 0.92, standard error of the estimate = 4.4, p < 0.05). However, acoustic quantification overestimated the radionuclide ejection fraction with rather wide limits of agreement (3.8% +/- 16.4%; bias +/- 2 SD). Thus echocardiographic automated border detection technique is a reasonably accurate method for on-line assessment of LV function.

Automated echocardiographic analysis of systemic ventricular performance in hypoplastic left heart syndrome

In patients with hypoplastic left heart syndrome, the right ventricle is positioned as the systemic ventricle. Monitoring its function is important but difficult because of its unusual position and geometry. The purposes of this study were to determine the feasibility of applying automatic echocardiographic measurements to the evaluation of right ventricular function in patients with hypoplastic left heart syndrome and to compare their function with that of normal subjects. Eleven patients with hypoplastic left heart syndrome (mean age 1.6 years) were evaluated with automatic border detection. Images were adequate if greater than 85% of the endocardium was being tracked. Systolic indexes were fractional area change and peak emptying rate; diastolic indexes were peak filling rate and the proportion of filling that occurred during rapid filling, diastasis, and atrial contraction. These data were compared with those in 18 normal control subjects. Three patients had completed stage 1 (aortic reconstruction, surgical shunt, and atrial septectomy), three patients had completed stage 2 (hemi-Fontan), and five patients had completed stage 3 (Fontan). Images were adequate in all cases. Fractional area change and peak emptying rate were significantly lower in the patients with hypoplastic left heart syndrome compared with control subjects (38% +/- 6% versus 61% +/- 10% and 3.5 +/- 0.9 versus 5.3 +/- 1.3 sec-1, respectively). Peak filling rate was significantly lower (4.1 +/- 1.3 versus 6.6 +/- 2.3 sec-1) and atrial contribution was significantly higher (18% +/- 7% versus 12% +/- 4%) in the patients. Automated echocardiographic analysis is feasible in assessing systemic ventricular function in children with hypoplastic left heart syndrome. These patients have abnormal systolic and diastolic function. Automatic border detection may be useful in detecting and monitoring systemic ventricular dysfunction as these patients progress through the surgical stages of repair.

Aortic elastic properties with transesophageal echocardiography with automated border detection: validation according to regional differences between proximal and distal descending thoracic aorta

We have previously described the use of transesophageal echocardiography with automated border detection to quantify regional aortic elastic properties. The purpose of this study was to validate this technique further by measuring regional variations of aortic elastic properties and comparing them with previously published data acquired by invasive methods. In nine anesthetized, closed-chest dogs, aortic pressure and lumenal area (transesophageal echocardiography with automated border detection) signals were recorded simultaneously at two aortic sites: just distal to the branching site of the left subclavian artery (proximal) and at the level of the diaphragm (distal). Instantaneous wall thickness was estimated by combining M-mode measurement of aortic end-diastolic thickness with instantaneous lumenal area. Data were acquired over a wide range of loading conditions, generated by inferior vena caval balloon occlusion. Aortic compliance per unit length, midwall radius, midwall stress, and incremental elastic modulus were computed. Aortic midwall radius and incremental elastic modulus values for proximal and distal aortic sites were compared at a common level of midwall stress. Compliance per unit length was higher in the proximal compared with the distal descending thoracic aorta (0.013 +/- 0.003 versus 0.008 +/- 0.003 cm2/mm Hg; mean +/- SD; p = 0.0011). Midwall radius was larger at the proximal location (0.76 +/- 0.07 cm versus 0.64 +/- 0.07 cm; p = 0.0001), whereas incremental elastic modulus was greater distally (0.799 +/- 0.052 dynes x 10(6)/cm2 versus 0.912 +/- 0.130 dynes x 10(6)/cm2; p = 0.02). Lower compliance values at the distal site of the descending thoracic aorta resulted from greater wall stiffness and a smaller radius. Transesophageal echocardiography with automated border detection provides reliable measurements of instantaneous aortic areas necessary for quantifying regional elastic properties.

Automatic backscatter analysis of regional left ventricular systolic function using color kinesis

Assessment of regional wall motion by 2-dimensional echocardiography can be performed by either semiquantitative wall motion scoring or by quantitative analysis. The former is subjective and requires expertise. Quantitative methods are too time-consuming for routine use in a busy clinical laboratory. Color kinesis is a new algorithm utilizing acoustic backscatter analysis. It provides a color encoded map of endocardial motion in real time. In each frame a new color layer is added; the thickness of the color beam represents endocardial motion during that frame. The end-systolic image has multiple color layers, representing regional and temporal heterogeneity of segmental motion. The purpose of this study was to validate the use of color kinesis for semiquantitative analysis of regional left ventricular systolic function and quantitatively in measurement of endocardial excursion. Semiquantitative wall motion scoring was performed in 18 patients using both 2-dimensional echo and color kinesis. Scoring was identical in 74% of segments; there was 84% agreement in definition of normal vs. abnormal. There was less interobserver variability in wall motion scoring using color kinesis. Endocardial excursion was quantified in 21 patients. 70% of the imaged segments were suitable for analysis. Correlation between 2-dimensional echocardiographic measurements and color kinesis was excellent, r = 0.87. The mean difference in excursion as measured by the 2 methods was -0.05 +/- 2.0 mm. In conclusion, color kinesis is a useful method for assessing regional contraction by displaying a color map of systolic endocardial excursion. This algorithm may improve the confidence and accuracy of assessment of segmental ventricular function by echocardiographic methods.

Echocardiographic quantification of regional left ventricular wall motion with color kinesis

BACKGROUND

Color kinesis is a new technology for the echocardiographic assessment of left ventricular wall motion based on acoustic quantification. This technique automatically detects endocardial motion in real time by using integrated backscatter data to identify pixel transitions from blood to tissue during systole on a frame-by-frame basis. In this study, we evaluated the feasibility and accuracy of quantitative segmental analysis of color kinesis images to provide objective evaluation of regional systolic endocardial motion.

METHODS AND RESULTS

Two-dimensional echocardiograms were obtained in the short-axis and apical four-chamber views in 20 normal subjects and 40 patients with regional wall motion abnormalities. End-systolic color overlays superimposed on the gray scale images were obtained with color kinesis to color encode left ventricular endocardial motion throughout systole on a frame-by-frame basis. These color-encoded images were divided into segments by use of custom software. In each segment, pixels of different colors were counted and displayed as stacked histograms reflecting the magnitude and timing of regional endocardial excursion. In normal subjects, histograms were found to be highly consistent and reproducible. The patterns of contraction obtained in normal subjects were used as a reference for the objective automated interpretation of regional wall motion abnormalities, defined as deviations from this pattern. The variability in the echocardiographic interpretation of wall motion between two experienced readers was similar to the diagnostic variability between the consensus of the two readers and the automated interpretation.

CONCLUSIONS

Color kinesis is a promising new tool that may be used clinically to improve the qualitative and quantitative evaluation of spatial and temporal aspects of global and regional wall motion. In this initial study, segmental analysis of color kinesis images provided accurate, automated, and quantitative diagnosis of regional wall motion abnormalities.

Beat-to-beat variability of left ventricular indexes measured by acoustic quantification: influence of heart rate and respiration--correlation with M-mode echocardiography

Influence of heart rate and respiration on beat-to-beat variability of left ventricular indexes measured by acoustic quantification was examined. These indexes were correlated with their counterparts measured by M-mode echocardiography. Parameters of left ventricular performance were recorded for 1 full minute in 43 children with a mean age of 5.9 +/- 3.9 years. Beat-to-beat variability was documented. The effect of respiration on such variability was examined in another 10 subjects. A wide range of heart rates and respiration did not show significant influence on the degree of variance among these parameters. The indexes measured correlated well with their counterparts measured by M-mode echocardiography. Acoustic quantification separated those with normal from those with abnormal left ventricular function with the same statistical significance as did M-mode echocardiography. A moderate degree of beat-to-beat variability occurs in acoustic quantification-derived left ventricular indexes. Heart rate variability and respiration do not influence the beat-to-beat variance of parameters of left ventricular performance measured with the acoustic quantification. Excellent correlation was documented between this technique and M-mode echocardiography.

Automatic border detection and three-dimensional reconstruction with echocardiography

This article reviews two important innovations in echocardiography resulting from the recent advances in the capabilities of microprocessors. The first, automatic endocardial border detection, has been implemented on computers contained entirely within echocardiograph machines and is gaining wide clinical use. The second, three-dimensional imaging, is currently under intense investigation and shows great promise for clinical application. It requires, however, further development of the specialized transducer apparatus necessary for image acquisition and the sophisticated computer-processing capability necessary for image reconstruction and display.

Left ventricular pressure-volume relations with transesophageal echocardiographic automated border detection: comparison with conductance-catheter technique

Pressure-volume relations are important means used to assess left ventricular (LV) contractility; however, on-line volume acquisition has been limited to the invasive conductance catheter. The objective was to compare simultaneous measures of LV volume by transesophageal echocardiographic automated border detection (ABD) and conductance catheter and their respective pressure-volume relations during steady state and alterations in preload and contractility. Seven dogs had placement of high-fidelity pressure and conductance catheters, a vena caval balloon occluder, and a transesophageal probe. An automated Simpson's rule volume algorithm was used from the transverse four-chamber view. Inotropic modulation was induced with dobutamine in four dogs and propranolol in three. Relative changes in ABD volume were linearly related to conductance volume at steady state with group mean r = 0.93 +/- 0.03, standard error of estimate (SEE) = 10 +/- 2%. Changes in end-diastolic volume, end-systolic volume, and stroke work with caval occlusion were also significantly correlated:r = 0.93 =/- 0.04, SEE = 3.6 ml; r = 0.89 +/- 0.04, SEE = 3.8 +/- 1.9 ml; and r = 0.86 +/- 0.05, SEE = 40 +/- 21 mJ, respectively. The overall bias was for absolute ABD volume to be less. End-systolic and maximal elastance values by ABD were significantly higher than by the conductance method; baseline group average 4.97 +/- 0.92 mm Hg/ml versus 2.70 +/- 1.15 mm Hg/ml and 6.63 +/- 1.66 mm Hg/ml versus 3.20 +/- 1.37 mm Hg/ml (p<0.05), respectively. However, the direction and relative magnitude of changes in elastance with inotropic modulation were similar.

Quantitative three-dimensional reconstruction of left ventricular volume with complete borders detected by acoustic quantification underestimates volume

Recently a new acoustic-quantification (AQ) technique has been developed to provide on-line automated border detection with an integrated backscatter analysis. Prior studies have largely correlated AQ areas with volumes without direct comparison of volumes for agreement. By using complete AQ-detected borders as the input to a validated method for three-dimensional echocardiographic (3DE) reconstruction, we can compare an entire cavity volume measured with the aid of AQ against a directly measured volume. This would also explore the possibility of applying AQ to 3DE reconstruction to reduce tracing time and enhance routine applicability. To compare reconstructed volumes with actual values in a stable standard allowing direct volume measurement, the left ventricles of 13 excised animal hearts were studied with a 3DE system that automatically combines two-dimensional (2D) images and their locations. Intersecting 2D views were obtained with conventional scanning and AQ imaging, with gains optimized to permit 3D reconstruction by detecting the most continuous AQ borders for each view, with maximal cavity size. Reconstruction was performed with manually traced central endocardial reflections and AQ-detected borders visually reproduced the left ventricular shapes; the AQ reconstructions, however, were consistently smaller. The reconstructed left ventricular (LV) volumes correlated well with actual values by both manual and AQ techniques (r = 0.93 and 0.88, with standard errors of 2.3 cc and 2.0 cc, p = not significant [NS]). Agreement with actual values was relatively close for the manually traced borders (y = 0.93x + 0.68, mean difference = -0.8 +/-2.2 cc). AQ-derived reconstructions consistently underestimated LV volume by 39 +/- 10% (y = 0.62x-0.09, mean difference = -7.8 +/- 3.0 cc, different from manually traced and actual volumes by analysis of variance [ANOVA], F = 69, p<0.00001). The AQ-detected threshold signal was displaced into the cavity, and volume between walls and false tendons was excluded, leading to underestimation, which increased with increasing cavity volume (r = 0.76). The AQ technique can therefore be applied to 3DE reconstruction, providing volumes that correlate well with directly measured values in a stable in vitro standard, minimizing observer decisions regarding manual border placement after image acquisition. However, when the complete borders needed for 3D reconstruction are used, absolute volumes are underestimated with current algorithms that integrate backscatter and displace the detected threshold into the ventricular cavity.

Automatic Border Detection to Assess Right Ventricular Function Following Surgical Treatment of Thromboembolic Pulmonary Hypertension

Automatic border detection (ABD) has been developed as a potentially useful means for evaluating ventricular function on line in an automatic fashion. Its success with tracking left ventricular function is established, but little is known about its ability to assess right ventricular (RV) function. Accordingly, 20 patients with severe pulmonary hypertension due to chronic thromboembolic disease underwent standard two-dimensional echocardiography and imaging with ABD before and after pulmonary thromboendarterectomy to correct pulmonary hypertension. ABD-derived results were compared to manually planimetered RV areas calculated from the apical four-chamber view. Doppler tricuspid regurgitant velocity fell significantly with surgery from 4.4 +/- 0.6 to 2.9 +/- 0.7 m/sec (P < 0.001). The mean values for RV areas derived by manual planimetry and ABD were similar, as was fractional area shortening, which improved significantly with surgery (manual 0.24 +/- 0.01 preoperative vs 0.31 +/- 0.11 postoperative, P < 0.05; and ABD 0.19 +/- 0.05 preoperative vs 0.32 +/- 0.15 postoperative, P < 0.001). There was, however, very little correlation between the individual values for ABD versus manually derived RV areas and fractional area shortening, with the best correlation being the RV end-diastolic areas after surgery (y = 0.684x + 7.9, r = 0.564, P = 0.01). These results demonstrate that both manually planimetered RV areas and those determined by ABD can adequately follow changes in ventricular function over time. However, variability within each technique may prevent direct comparison of the absolute values of the two techniques. (ECHOCARDIOGRAPHY, Volume 13, March 1996)

Improvement of left ventricular diastolic dysfunction in hypertensive patients 1 month after ACE inhibition therapy: evaluation by ultrasonic automated boundary detection

The purpose of this study was to detect any improvement in left ventricular diastolic dysfunction in hypertensive patients 1 month after cilazapril therapy. Twenty-three patients, 5 men and 18 women (mean age, 53.52 +/- 9.10 years), with mild or moderate hypertension (160 +/- 13/98 +/- 10 mm Hg), and free of other cardiac or systemic diseases, were studied using ultrasonic automated boundary detection (ABD) and pulsed Doppler echocardiography, before and 1 month after a daily dose of 2.5 mg of cilazapril. The following new ABD diastolic indices were determined: the time rate of area change in early diastole (dA/dt)E, that in late diastole (dA/dt)A, and their ratio (dA/dt)E/(dA/dt)A, while Doppler transmitral flow measurements of left ventricular diastolic filling were also simultaneously recorded. The ABD results showed left ventricular diastolic dysfunction (LVDD) in 9 of 23 patients (39%) compared with the ABD values of 12 normal volunteers. Neither method revealed any significant difference before and after treatment in the patient group as a whole. However, in the group of 9 patients with diastolic dysfunction, the ABD ratio (dA/dt)E/(dA/dt)A was significantly improved after cilazapril therapy (1.20 +/- 0.21 versus 1.41 +/- 0.17; P < 0.05). We concluded that a large percentage (39%) of patients with mild or moderate hypertension had reduced diastolic performance of the left ventricle at a stage of the disease when systolic dysfunction and/or hypertrophy were not evident. Significant improvement of diastolic dysfunction in hypertensive patients could be detected by the proposed ABD new diastolic indices 1 month after cilazapril therapy. In conclusion, automatic boundary detection should be a useful non-invasive modality for the early diagnosis of left ventricular diastolic dysfunction, as well as early recognition of its improvement.

The clinical utility of automatic boundary detection for the determination of left ventricular volume: a comparison with conventional off-line echocardiographic quantification

The aim of this study was to compare measurements of echocardiographic volume with an on-line automatic boundary detection imaging system with those of a conventional off-line method for routine clinical studies. Automatic boundary detection imaging shows promise as a rapid, on-line method for quantitating left ventricular volumes by echocardiography. However, there is little information about the role of automatic boundary detection for routine clinical studies. Ninety-seven patients with a variety of clinical diseases who were referred for clinical transthoracic echocardiographic evaluation were studied in apical four-chamber and two-chamber imaging planes. End-diastolic volume, end-systolic volume, and ejection fraction obtained with automatic boundary detection images were compared with those of conventional off-line analysis. Segmental endocardial definition and border tracking were evaluated on all automatic boundary detection images. Left ventricular end-diastolic volumes obtained by automatic boundary detection correlated well but were systematically under-estimated compared with off-line analysis for the apical two-chamber (r = 0.83; underestimation = 42 +/- 33 ml; p < 0.05) and four-chamber views (r = 0.83; underestimation = 43 +/- 31 ml; p < 0.05). Left ventricular end-systolic volumes also correlated well but were underestimated by automatic boundary detection for the apical two-chamber (r = 0.83; underestimation = 14 +/- 26 ml; p < 0.05) and four-chamber views (r = 0.83; underestimation = 18 +/- 24 ml; p < 0.05). Ejection fraction was not predicted accurately for the entire study population (n = 97). However, for patients with complete endocardial definition (n = 32), automatic boundary detection accurately predicted ejection fraction with no systematic error compared with manually traced images for both the apical two-chamber (r = 0.86; p < 0.05) and four-chamber (r = 0.82; p < 0.05) views. Segmental analysis of endocardial tracking revealed significantly better tracking of the septal and lateral walls compared with other regions (p < 0.05). End-diastolic and end-systolic volumes determined by automatic boundary detection correlate well but underestimate volume compared with conventional off-line analysis. However, ejection fraction compares favorably for the two methods when there is complete endocardial definition.

Improved quantification of left ventricular function by applying signal averaging to echocardiographic acoustic quantification

The acoustic quantification technique for on-line detection of endocardial boundaries currently provides continuous left ventricular area or volume signals and beat-to-beat ejection fraction. However, the distortion of individual waveforms by noise results in a wide beat-to-beat variability in these parameters. We developed an automated algorithm for the evaluation of left ventricular function by averaging acoustic quantification signals. End-diastolic and end-systolic area, stroke area, and fractional area change are measured directly from the average waveform. Peak ejection and peak filling rates and time to peak filling rate are obtained from its time derivative. Area signals obtained from eight normal subjects were used to evaluate the performance of this algorithm. Parameters of left ventricular function obtained with the automated algorithm were highly consistent and in excellent agreement with those obtained by repeated manual operator-dependent selections. This algorithm provides a fast and easy method for noise reduction in acoustic quantification signals, which significantly improves the noninvasive assessment of left ventricular function.

Echocardiographic automatic boundary detection to measure left atrial function after the maze procedure

Automatic boundary detection (ABD) is a new echocardiographic modality providing continuous on-line measurements of cavitary area throughout the cardiac cycle. The maze procedure is a new surgical intervention designed to restore sinus rhythm and mechanical atrial contraction as a definitive treatment for patients with atrial fibrillations for whom medical therapy has failed. To evaluate whether ABD may define left atrial function in patients after the maze procedure, we obtained pulsed Doppler recordings of mitral inflow velocity and echocardiographic ABD in 25 patients, 6 +/- 2 months after the maze procedure. We measured the left atrial end-systolic cavitary area, mid-diastolic area before atrial contraction, and end-diastolic area (in square centimeters). Left atrial contraction by Doppler was compared with that derived by ABD in patients who underwent the maze procedure and control subjects (n = 13), both qualitatively and quantitatively (atrial filling fraction vs active atrial contraction [ABD] where atrial contraction (in percent) = (mid-diastolic area - end-diastolic area) x 100/(end-systolic area - end-diastolic area in percent]). Restoration of atrial contraction after the maze procedure was detected by Doppler in 19 patients (76%) and by ABD in 21 patients (84%). The atrial filling fraction was 19 +/- 4% in patients compared with values of 34% +/- 8% in control subjects (p < 0.001). By ABD atrial contraction was 20% +/- 6% in patients whereas control subjects exhibited values of 41% +/- 14% p < 0.001). The Doppler-derived atrial filling fraction and ABD-derived atrial contraction were closely correlated (r = 0.91; p < 0.001; y = 0.59x + 8.6). Thus Doppler techniques complemented by ABD provide direct quantitative indexes of left atrial function throughout the cardiac cycle. Although left atrial contraction and filling are reduced after the maze procedure, left atrial function is restored in most patients with a history of atrial fibrillation, and echocardiographic ABD is a sensitive technique for its detection.