Spatial-temporal ultrasound imaging of residual cavitation bubbles around a fluid-tissue interface in histotripsy

Minimum variance beamforming combined with covariance matrix-based adaptive weighting for medical ultrasound imaging

BACKGROUND

The minimum variance (MV) beamformer can significantly improve the image resolution in ultrasound imaging, but it has limited performance in noise reduction. We recently proposed the covariance matrix-based statistical beamforming (CMSB) for medical ultrasound imaging to reduce sidelobes and incoherent clutter.

METHODS

In this paper, we aim to improve the imaging performance of the MV beamformer by introducing a new pixel-based adaptive weighting approach based on CMSB, which is named as covariance matrix-based adaptive weighting (CMSAW). The proposed CMSAW estimates the mean-to-standard-deviation ratio (MSR) of a modified covariance matrix reconstructed by adaptive spatial smoothing, rotary averaging, and diagonal reducing. Moreover, adaptive diagonal reducing based on the aperture coherence is introduced in CMSAW to enhance the performance in speckle preservation.

RESULTS

The proposed CMSAW-weighted MV (CMSAW-MV) was validated through simulation, phantom experiments, and in vivo studies. The phantom experimental results show that CMSAW-MV obtains resolution improvement of 21.3% and simultaneously achieves average improvements of 96.4% and 71.8% in average contrast and generalized contrast-to-noise ratio (gCNR) for anechoic cyst, respectively, compared with MV. in vivo studies indicate that CMSAW-MV improves the noise reduction performance of MV beamformer.

CONCLUSION

Simulation, experimental, and in vivo results all show that CMSAW-MV can improve resolution and suppress sidelobes and incoherent clutter and noise. These results demonstrate the effectiveness of CMSAW in improving the imaging performance of MV beamformer. Moreover, the proposed CMSAW with a computational complexity of [Formula: see text] has the potential to be implemented in real time using the graphics processing unit.

Acoustic characterization of cavitation intensity: A review

Cavitation intensity is used to describe the activity of cavitation, and several methods are developed to identify the intensity of cavitation. This work aimed to provide an overview and discussion of the several existing characterization methods for cavitation intensity, three acoustic approaches for charactering cavitation were discussed in detail. It was showed that cavitation noise spectrum is too complex and there are some differences and disputes on the characterization of cavitation intensity by cavitation noise. In this review, we recommended a total cavitation noise intensity estimated via the integration of real cavitation noise spectrum over full frequency domain instead of artificially adding inaccurate filtering processing.

Observation and modulation of the dissolution of histotripsy-induced bubble clouds with high-frame rate plane wave imaging

Focused ultrasound therapies are a noninvasive means to ablate tissue. Histotripsy utilizes short ultrasound pulses with sufficient tension to nucleate bubble clouds that impart lethal strain to the surrounding tissues. Tracking bubble cloud dissolution between the application of histotripsy pulses is critical to ensure treatment efficacy. In this study, plane wave B-mode imaging was employed to monitor bubble cloud motion and grayscale at frame rates up to 11.25 kHz. Minimal changes in the area or position of the bubble clouds were observed 50 ms post excitation. The bubble cloud grayscale was observed to decrease with the square root of time, indicating a diffusion-driven process. These results were qualitatively consistent with an analytic model of gas diffusion during the histotripsy process. Finally, the rate of bubble cloud dissolution was found to be dependent on the output of the imaging pulse, indicating an interaction between the bubble cloud and imaging parameters. Overall, these results highlight the utility of plane wave B-mode imaging for monitoring histotripsy bubble clouds.

MRI-guided transurethral insonation of silica-shell phase-shift emulsions in the prostate with an advanced navigation platform

PURPOSE

In this study, the efficacy of transurethral prostate ablation in the presence of silica-shell ultrasound-triggered phase-shift emulsions (sUPEs) doped with MR contrast was evaluated. The influence of sUPEs on MR imaging assessment of the ablation zone was also investigated.

METHODS

sUPEs were doped with a magnetic resonance (MR) contrast agent, Gd₂ O₃ , to assess ultrasound transition. Injections of saline (sham), saline and sUPEs alone, and saline and sUPEs with Optison microbubbles were performed under guidance of a prototype interventional MRI navigation platform in a healthy canine prostate. Treatment arms were evaluated for differences in lesion size, T₁  contrast, and temperature. In addition, non-perfused areas (NPAs) on dynamic contrast-enhanced (DCE) MRI, 55°C isotherms, and areas of 240 cumulative equivalent minutes at 43°C (CEM₄₃ ) dose or greater computed from MR thermometry were measured and correlated with ablated areas indicated by histology.

RESULTS

For treatment arms including sUPEs, the computed correlation coefficients between the histological ablation zone and the NPA, 55°C isotherm, and 240 CEM₄₃ area ranged from 0.96-0.99, 0.98-0.99, and 0.91-0.99, respectively. In the absence of sUPEs, the computed correlation coefficients between the histological ablation zone and the NPA, 55°C isotherm, and 240 CEM₄₃ area were 0.69, 0.54, and 0.50, respectively. Across all treatment arms, the areas of thermal tissue damage and NPAs were not significantly different (P = 0.47). Areas denoted by 55°C isotherms and 240 CEM₄₃ dose boundaries were significantly larger than the areas of thermal damage, again for all treatment arms (P = 0.009 and 0.003, respectively). No significant differences in lesion size, T₁ contrast, or temperature were observed between any of the treatment arms (P > 0.0167). Lesions exhibiting thermal fixation on histological analysis were present in six of nine insonations involving sUPE injections and one of five insonations involving saline sham injections. Significantly larger areas (P = 0.002), higher temperatures (P = 0.004), and more frequent ring patterns of restricted diffusion on ex vivo diffusion-weighted imaging (P = 0.005) were apparent in lesions with thermal fixation.

CONCLUSIONS

T₁ contrast suggesting sUPE transition was not evident in sUPE treatment arms. The use of MR imaging metrics to predict prostate ablation was not diminished by the presence of sUPEs. Lesions generated in the presence of sUPEs exhibited more frequent thermal fixation, though there were no significant changes in the ablation areas when comparing arms with and without sUPEs. Thermal fixation corresponded to some qualitative imaging features.

Ultrasonic sector imaging using plane wave synthetic focusing with a convex array transducer

Synthetic transmit focusing (STF) methods using unfocused waves or plane waves (PWs) have recently been investigated extensively. However, STF using PWs with a convex array (PWSTF-CA) has not been rigorously studied for high-resolution sector imaging. In this paper, the authors suggest an analytical model for accurate characterization of the spatial beam patterns of PWSTF-CA using a large range of either uniformly or non-uniformly distributed PW angles. On the basis of the model, a frame-based PWSTF-CA approach with non-uniform PW angles is suggested to achieve superior image quality at a higher frame rate than conventional transmit focusing (CTF). The analytical model can also be used for optimal selection of a set of PW angles to scan the entire sectorial field of view and its subsets employed for STF at each imaging point. The authors also investigate how to select transmit subarrays for each of the PWs to obtain the best spatial resolution. A theoretical analysis and simulations are conducted for the verification of the analytical model and the optimal utilization strategy of PWSTF-CA. The results indicate that the PWSTF-CA improves not only the frame rate but also the contrast, signal-to-noise ratio, and resolution compared with the CTF, as in the case of PWSTF with linear arrays.

Nakagami-m parametric imaging for characterization of thermal coagulation and cavitation erosion induced by HIFU

Nowadays, both thermal and mechanical ablation techniques of HIFU associated with cavitation have been developed for noninvasive treatment. A specific challenge for the successful clinical implementation of HIFU is to achieve real-time imaging for the evaluation and determination of therapy outcomes such as necrosis or homogenization. Ultrasound Nakagami-m parametric imaging highlights the degrading shadowing effects of bubbles and can be used for tissue characterization. The aim of this study is to investigate the performance of Nakagami-m parametric imaging for evaluating and differentiating thermal coagulation and cavitation erosion induced by HIFU. Lesions were induced in basic bovine serum albumin (BSA) phantoms and ex vivo porcine livers using a 1.6 MHz single-element transducer. Thermal and mechanical lesions induced by two types of HIFU sequences respectively were evaluated using Nakagami-m parametric imaging and ultrasound B-mode imaging. The lesion sizes estimated using Nakagami-m parametric imaging technique were all closer to the actual sizes than those of B-mode imaging. The p-value obtained from the t-test between the mean m values of thermal coagulation and cavitation erosion was smaller than 0.05, demonstrating that the m values of thermal lesions were significantly different from that of mechanical lesions, which was confirmed by ex vivo experiments and histologic examination showed that different changes result from HIFU exposure, one of tissue dehydration resulting from the thermal effect, and the other of tissue homogenate resulting from mechanical effect. This study demonstrated that Nakagami-m parametric imaging is a potential real-time imaging technique for evaluating and differentiating thermal coagulation and cavitation erosion.

Design and Characterization of an Acoustically and Structurally Matched 3-D-Printed Model for Transcranial Ultrasound Imaging

For investigating human transcranial ultrasound imaging (TUI) through the temporal bone, an intact human skull is needed. Since it is complex and expensive to obtain one, it requires that experiments are performed without excision or abrasion of the skull. Besides, to mimic blood circulation for the vessel target, cellulose tubes generally fit the vessel simulation with straight linear features. These issues, which limit experimental studies, can be overcome by designing a 3-D-printed skull model with acoustic and dimensional properties that match a real skull and a vessel model with curve and bifurcation. First, the optimal printing material which matched a real skull in terms of the acoustic attenuation coefficient and sound propagation velocity was identified at 2-MHz frequency, i.e., 7.06 dB/mm and 2168.71 m/s for the skull while 6.98 dB/mm and 2114.72 m/s for the printed material, respectively. After modeling, the average thickness of the temporal bone in the printed skull was about 1.8 mm, while it was to 1.7 mm in the real skull. Then, a vascular phantom was designed with 3-D-printed vessels of low acoustic attenuation (0.6 dB/mm). It was covered with a porcine brain tissue contained within a transparent polyacrylamide gel. After characterizing the acoustic consistency, based on the designed skull model and vascular phantom, vessels with inner diameters of 1 and 0.7 mm were distinguished by resolution enhanced imaging with low frequency. Measurements and imaging results proved that the model and phantom are authentic and viable alternatives, and will be of interest for TUI, high intensity focused ultrasound, or other therapy studies.

Passive acoustic mapping of cavitation using eigenspace-based robust Capon beamformer in ultrasound therapy

Pulse-echo imaging technique can only play a role when high intensity focused ultrasound (HIFU) is turned off due to the interference between the primary HIFU signal and the transmission pulse. Passive acoustic mapping (PAM) has been proposed as a tool for true real-time monitoring of HIFU therapy. However, the most-used PAM algorithm based on time exposure acoustic (TEA) limits the quality of cavitation image. Recently, robust Capon beamformer (RCB) has been used in PAM to provide improved resolution and reduced artifacts over TEA-based PAM, but the presented results have not been satisfactory. In the present study, we applied an eigenspace-based RCB (EISRCB) method to further improve the PAM image quality. The optimal weighting vector of the proposed method was found by projecting the RCB weighting vector onto the desired vector subspace constructed from the eigenstructure of the covariance matrix. The performance of the proposed PAM was validated by both simulations and in vitro histotripsy experiments. The results suggested that the proposed PAM significantly outperformed the conventionally used TEA and RCB-based PAM. The comparison results between pulse-echo images of the residual bubbles and cavitation images showed the potential of our proposed PAM in accurate localization of cavitation activity during HIFU therapy.

Post Hoc Analysis of Passive Cavitation Imaging for Classification of Histotripsy-Induced Liquefaction in Vitro

Histotripsy utilizes focused ultrasound to generate bubble clouds for transcutaneous tissue liquefaction. Bubble activity maps are under development to provide image guidance and monitor treatment progress. The aim of this paper was to investigate the feasibility of using plane wave B-mode and passive cavitation images to be used as binary classifiers of histotripsy-induced liquefaction. Prostate tissue phantoms were exposed to histotripsy pulses over a range of pulse durations (5- ) and peak negative pressures (12-23 MPa). Acoustic emissions were recorded during the insonation and beamformed to form passive cavitation images. Plane wave B-mode images were acquired following the insonation to detect the hyperechoic bubble cloud. Phantom samples were sectioned and stained to delineate the liquefaction zone. Correlation between passive cavitation and plane wave B-mode images and the liquefaction zone was assessed using receiver operating characteristic (ROC) curve analysis. Liquefaction of the phantom was observed for all the insonation conditions. The area under the ROC (0.94 versus 0.82), accuracy (0.90 versus 0.83), and sensitivity (0.81 versus 0.49) was greater for passive cavitation images relative to B-mode images ( ) along the azimuth of the liquefaction zone. The specificity was greater than 0.9 for both imaging modalities. These results demonstrate a stronger correlation between histotripsy-induced liquefaction and passive cavitation imaging compared with the plane wave B-mode imaging, albeit with limited passive cavitation image range resolution.

Pulse-Inversion Subharmonic Ultrafast Active Cavitation Imaging in Tissue Using Fast Eigenspace-Based Adaptive Beamforming and Cavitation Deconvolution

Pulse-inversion subharmonic (PISH) imaging can display information relating to pure cavitation bubbles while excluding that of tissue. Although plane-wave-based ultrafast active cavitation imaging (UACI) can monitor the transient activities of cavitation bubbles, its resolution and cavitation-to-tissue ratio (CTR) are barely satisfactory but can be significantly improved by introducing eigenspace-based (ESB) adaptive beamforming. PISH and UACI are a natural combination for imaging of pure cavitation activity in tissue; however, it raises two problems: 1) the ESB beamforming is hard to implement in real time due to the enormous amount of computation associated with the covariance matrix inversion and eigendecomposition and 2) the narrowband characteristic of the subharmonic filter will incur a drastic degradation in resolution. Thus, in order to jointly address these two problems, we propose a new PISH-UACI method using novel fast ESB (F-ESB) beamforming and cavitation deconvolution for nonlinear signals. This method greatly reduces the computational complexity by using F-ESB beamforming through dimensionality reduction based on principal component analysis, while maintaining the high quality of ESB beamforming. The degraded resolution is recovered using cavitation deconvolution through a modified convolution model and compressive deconvolution. Both simulations and in vitro experiments were performed to verify the effectiveness of the proposed method. Compared with the ESB-based PISH-UACI, the entire computation of our proposed approach was reduced by 99%, while the axial resolution gain and CTR were increased by 3 times and 2 dB, respectively, confirming that satisfactory performance can be obtained for monitoring pure cavitation bubbles in tissue erosion.

Coherence factor and Wiener postfilter in synthetic aperture ultrasound imaging

The coherence factor (CF) and Wiener postfilter methods have been proposed as effective approaches for reducing the output noise of the delay-and-sum (DAS) beamformer in ultrasound imaging. The theoretical framework between them was also established. However, past researches about the CF and Wiener postfilter methods mainly focused on the summation of an array signal. This paper analyzes the CF and Wiener postfilter in the synthetic aperture (SA) imaging mode, where two-dimensional echo data are recorded. Different CF definitions in the SA imaging are first given and the corresponding Wiener postfilter methods are then proposed, including a Wiener postfilter especially for the SA imaging, named as Wiener_SA. The performances of different CF and Wiener postfilter methods were evaluated on both simulated and experimental SA data. Results showed that the proposed Wiener_SA outperformed the other Wiener postfilters in reducing the sidelobe noise level. It obtained the highest contrast ratio among the Wiener postfilter methods, which was even higher than some of the CF methods. Meanwhile it could benefit a much higher contrast-to-noise ratio than those CF methods with further suppression of incoherent noises. Consequently, the Wiener_SA is believed to be a promising approach in enhancing the SA image quality.

Wavelet-transform-based active imaging of cavitation bubbles in tissues induced by high intensity focused ultrasound

Cavitation detection and imaging are essential for monitoring high-intensity focused ultrasound (HIFU) therapies. In this paper, an active cavitation imaging method based on wavelet transform is proposed to enhance the contrast between the cavitation bubbles and surrounding tissues. The Yang-Church model, which is a combination of the Keller-Miksis equation with the Kelvin-Voigt equation for the pulsations of gas bubbles in simple linear viscoelastic solids, is utilized to construct the bubble wavelet. Experiments with porcine muscles demonstrate that image quality is associated with the initial radius of the bubble wavelet and the scale. Moreover, the Yang-Church model achieves a somewhat better performance compared with the Rayleigh-Plesset-Noltingk-Neppiras-Poritsky model. Furthermore, the pulse inversion (PI) technique is combined with bubble wavelet transform to achieve further improvement. The cavitation-to-tissue ratio (CTR) of the best tissue bubble wavelet transform (TBWT) mode image is improved by 5.1 dB compared with that of the B-mode image, while the CTR of the best PI-based TBWT mode image is improved by 7.9 dB compared with that of the PI-based B-mode image. This work will be useful for better monitoring of cavitation in HIFU-induced therapies.

High-contrast active cavitation imaging technique based on multiple bubble wavelet transform

In this study, a unique method that combines the ultrafast active cavitation imaging technique with multiple bubble wavelet transform (MBWT) for improving cavitation detection contrast was presented. The bubble wavelet was constructed by the modified Keller-Miksis equation that considered the mutual effect among bubbles. A three-dimensional spatial model was applied to simulate the spatial distribution of multiple bubbles. The effects of four parameters on the signal-to-noise ratio (SNR) of cavitation images were evaluated, including the following: initial radii of bubbles, scale factor in the wavelet transform, number of bubbles, and the minimum inter-bubble distance. And the other two spatial models and cavitation bubble size distributions were introduced in the MBWT method. The results suggested that in the free-field experiments, the averaged SNR of images acquired by the MBWT method was improved by 7.16 ± 0.09 dB and 3.14 ± 0.14 dB compared with the values of images acquired by the B-mode and single bubble wavelet transform (SBWT) methods. In addition, in the tissue experiments, the averaged cavitation-to-tissue ratio of cavitation images acquired by the MBWT method was improved by 4.69 ± 0.25 dB and 1.74± 0.29 dB compared with that of images acquired by B-mode and SBWT methods.

Spatial-temporal three-dimensional ultrasound plane-by-plane active cavitation mapping for high-intensity focused ultrasound in free field and pulsatile flow

Cavitation plays important roles in almost all high-intensity focused ultrasound (HIFU) applications. However, current two-dimensional (2D) cavitation mapping could only provide cavitation activity in one plane. This study proposed a three-dimensional (3D) ultrasound plane-by-plane active cavitation mapping (3D-UPACM) for HIFU in free field and pulsatile flow. The acquisition of channel-domain raw radio-frequency (RF) data in 3D space was performed by sequential plane-by-plane 2D ultrafast active cavitation mapping. Between two adjacent unit locations, there was a waiting time to make cavitation nuclei distribution of the liquid back to the original state. The 3D cavitation map equivalent to the one detected at one time and over the entire volume could be reconstructed by Marching Cube algorithm. Minimum variance (MV) adaptive beamforming was combined with coherence factor (CF) weighting (MVCF) or compressive sensing (CS) method (MVCS) to process the raw RF data for improved beamforming or more rapid data processing. The feasibility of 3D-UPACM was demonstrated in tap-water and a phantom vessel with pulsatile flow. The time interval between temporal evolutions of cavitation bubble cloud could be several microseconds. MVCF beamformer had a signal-to-noise ratio (SNR) at 14.17dB higher, lateral and axial resolution at 2.88times and 1.88times, respectively, which were compared with those of B-mode active cavitation mapping. MVCS beamformer had only 14.94% time penalty of that of MVCF beamformer. This 3D-UPACM technique employs the linear array of a current ultrasound diagnosis system rather than a 2D array transducer to decrease the cost of the instrument. Moreover, although the application is limited by the requirement for a gassy fluid medium or a constant supply of new cavitation nuclei that allows replenishment of nuclei between HIFU exposures, this technique may exhibit a useful tool in 3D cavitation mapping for HIFU with high speed, precision and resolution, especially in a laboratory environment where more careful analysis may be required under controlled conditions.