Dual-spectra ultrasonography: an attenuation-compensating technique for myocardial perfusion analysis

Attenuation correction in ultrasound contrast agent imaging: elementary theory and preliminary experimental evaluation

Progress in imaging and quantification of tissue perfusion using ultrasound (US) and microbubble contrast agents has been undermined by the lack of an effective automatic attenuation correction technique. In this article, an elementary model of the US attenuation processes for microbubble contrast enhanced imaging is developed. In the model, factors such as nonlinear bubble scattering, nonlinear attenuation, attenuation to both fundamental and harmonic and the US beam profile are considered. Methods are proposed for fast formation of images with automatic attenuation correction based on the model. In the proposed method, linear tissue echoes are extracted and filtered and then used to compensate for the attenuation in nonlinear bubble echoes at the same location to produce quantities that are a truer representation of bubble concentration. The technique does not require additional measurements and can be implemented in real time. Preliminary experiments on laboratory phantoms consisting of bubbles and tissue-mimicking materials are presented and the effectiveness of the proposed method is supported by improvements in image quality compared with unprocessed data. This development is an important step towards real-time quantitative contrast US imaging.

Parametric detection and measurement of perfusion defects in attenuated contrast echocardiographic images

OBJECTIVE

Attenuation of radio frequency (RF) signals limits the use of contrast echocardiography. The harmonic-to-fundamental ratio (HFR) of the RF signals compensates for attenuation. We tested whether HFR analysis measures the left ventricular nonperfused area under simulated experimental attenuation.

METHODS

Radio frequency image data from short axis systolic projections were obtained from 11 open-chest dogs with left anterior descending or left circumflex coronary artery occlusion followed by left atrial bolus injection of a perflutren microbubble contrast agent. Clinical attenuation was simulated by calibrated silicone pads interposed between the epicardial surface and the transducer to induce mild (7-dB) and severe (14-dB) reduction of the backscattered RF signals. Harmonic-to-fundamental ratio values were calculated for each image pixel for 0-, 7-, and 14-dB attenuation conditions and reproducibly showed a "perfused area" and a "nonperfused area." A reference nonperfused area was obtained by manual delineation in high-quality contrast scans.

RESULTS

Correlations of the HFR-detected and manually outlined perfusion defect areas were R = 0.92 for 0 dB, R = 0.94 for 7 dB, and R = 0.90 for 14 dB; the mean difference was less than 0.36 cm(2) (negligible) in all 3 attenuation settings. Conclusions. Attenuation compensation by our HFR method allows precise measurement of myocardial perfusion defect areas in contrast scans with simulated high level of attenuation.

Parametric harmonic-to-fundamental ratio contrast echocardiography: a novel approach to identification and accurate measurement of left ventricular area under variable levels of ultrasound signal attenuation

OBJECTIVES

We introduced a harmonic-to-fundamental ratio (HFR) of the radiofrequency (RF) signals that reduces confounding effects of attenuation. We studied whether HFR analysis of RF signals received from contrast microbubbles allows accurate measurement of the left ventricular (LV) cavity area under varying levels of attenuation.

BACKGROUND

Attenuation is a fundamental problem in ultrasound imaging and limits the use of clinical echocardiography.

METHODS

RF data from short axis systolic and diastolic scans were obtained from 14 open-chest dogs following left-atrial bolus of Optison. Attenuation was induced by interposed silicone pads calibrated to induce 7dB or 14dB reductions of the backscattered RF signal. RF images were reconstructed from the RF signals, HFR values calculated for each image pixel for 0dB, 7dB and 14dB attenuation conditions, and LV area obtained by summation of "LV cavity pixels". A reference LV cavity area was obtained from endocardial border tracings in enhanced scans by experts.

RESULTS

Correlation of the HFR-defined and reference areas at systole was R=0.95, R=0.94, and R=0.91 for 0dB, 7dB and 14dB levels of attenuation, respectively, and at diastole was R=0.95 for 0dB, 7dB and 14dB levels of attenuation. The mean difference from both systolic and diastolic values was <1.45 cm(2) (i.e. negligible) in all attenuation settings.

CONCLUSION

Our novel HFR method supports precise measurement of the LV cavity area in contrast images with simulated high attenuation of ultrasound signals.

Spectral normalization for ultrasonic contrast microbubble detection

Ultrasonic contrast agents consisting of microbubbles are used to assess tissue perfusion. The microbubbles are highly reflective and nonlinear and thus produce harmonics that are stronger than those from tissues. However, the magnitude of harmonic signals resulting from a region with microbubbles also depends on the acoustic pressure of incident ultrasound and the attenuation of intervening tissues in the ultrasound path. Therefore, the harmonic magnitude, as used in traditional harmonic imaging, may not be a reliable indicator of the presence or absence of microbubbles, and hence, tissue perfusion. To compensate for these effects, we present two parameters defined as the ratio of the harmonic to the fundamental component (HFR) and the ratio of the harmonic to squared fundamental (HSFR). A simplified model is used to illustrate the usefulness of these two parameters. Experiments show that both parameters improve detection of microbubbles and that HSFR performs better than HFR.

Radio frequency dual-spectra analysis of regional myocardial perfusion: Comparison with harmonic densitometric method

Our objective was to compare harmonic-to-fundamental frequency ratio peak (HFRp) analysis with conventional harmonic gray-scale densitometric analysis on the basis of ability to eliminate heterogeneity in ultrasound signals. Broadband radio frequency and harmonic data were obtained by using intermittent short-axis scans in 10 open-chest pigs before and during infusion of contrast microbubbles. HFRp and gray-scale intensity values were measured in 6 segments of left ventricular myocardium. In baseline images, the influence of anisotropy on HFRp values was significantly less than that in gray-scale intensities. In perfusion assessment with subtraction, contrast heterogeneity in HFRp values was significantly smaller than that in gray-scale intensities. The increase in HFRp values after subtraction was significantly greater than that in gray-scale intensities in lateral (P <.001), posterior (P <.0001), and inferior (P <.01) myocardium. HFRp analysis can compensate for baseline myocardial anisotropy and regional contrast heterogeneity. With background subtraction, HFRp analysis allows better quantification of myocardial perfusion.