Dongwoon Hyun

Lesion Detectability in Diagnostic Ultrasound with Short-Lag Spatial Coherence Imaging

We demonstrate a novel imaging technique, named short-lag spatial coherence (SLSC) imaging, which uses short distance (or lag) values of the coherence function of backscattered ultrasound to create images. Simulations using Field II are used to demonstrate the detection of lesions of varying sizes and contrasts with and without acoustical clutter in the backscattered data. B-mode and SLSC imaging are shown to be nearly equivalent in lesion detection, based on the contrast-to-noise ratio (CNR) of the lesion, in noise-free conditions. The CNR of the SLSC image, however, can be adjusted to achieve an optimal value at the expense of image smoothness and resolution. In the presence of acoustic clutter, SLSC imaging yields significantly higher CNR than B-mode imaging and maintains higher image quality than B-mode with increasing noise. Compression of SLSC images is shown to be required under high-noise conditions but is unnecessary under no- and low-noise conditions. SLSC imaging is applied to in vivo imaging of the carotid sheath and demonstrates significant gains in CNR as well as visualization of arterioles in the carotid sheath. SLSC imaging has a potential application to clutter rejection in ultrasonic imaging.

Short-lag spatial coherence imaging on matrix arrays, Part 1: Beamforming methods and simulation studies

Short-lag spatial coherence (SLSC) imaging is a beamforming technique that has demonstrated improved imaging performance compared with conventional B-mode imaging in previous studies. Thus far, the use of 1-D arrays has limited coherence measurements and SLSC imaging to a single dimension. Here, the SLSC algorithm is extended for use on 2-D matrix array transducers and applied in a simulation study examining imaging performance as a function of subaperture configuration and of incoherent channel noise. SLSC images generated with a 2-D array yielded superior contrast-to-noise ratio (CNR) and texture SNR measurements over SLSC images made on a corresponding 1-D array and over B-mode imaging. SLSC images generated with square subapertures were found to be superior to SLSC images generated with subapertures of equal surface area that spanned the whole array in one dimension. Subaperture beamforming was found to have little effect on SLSC imaging performance for subapertures up to 8 × 8 elements in size on a 64 × 64 element transducer. Additionally, the use of 8 × 8, 4 × 4, and 2 × 2 element subapertures provided 8, 4, and 2 times improvement in channel SNR along with 2640-, 328-, and 25-fold reduction in computation time, respectively. These results indicate that volumetric SLSC imaging is readily applicable to existing 2-D arrays that employ subaperture beamforming.

Efficient Strategies for Estimating the Spatial Coherence of Backscatter

The spatial coherence of ultrasound backscatter has been proposed to reduce clutter in medical imaging, to measure the anisotropy of the scattering source, and to improve the detection of blood flow. These techniques rely on correlation estimates that are obtained using computationally expensive strategies. In this paper, we assess the existing spatial coherence estimation methods and propose three computationally efficient modifications: a reduced kernel, a downsampled receive aperture, and the use of an ensemble correlation coefficient. The proposed methods are implemented in simulation and in vivo studies. Reducing the kernel to a single sample improved computational throughput and improved axial resolution. Downsampling the receive aperture was found to have negligible effect on estimator variance, and improved computational throughput by an order of magnitude for a downsample factor of 4. The ensemble correlation estimator demonstrated lower variance than the currently used average correlation. Combining the three methods, the throughput was improved 105-fold in simulation with a downsample factor of 4- and 20-fold in vivo with a downsample factor of 2.

Visualization of Small-Diameter Vessels by Reduction of Incoherent Reverberation With Coherent Flow Power Doppler

Power Doppler (PD) imaging is a widely used technique for flow detection. Despite the wide use of Doppler ultrasound, limitations exist in the ability of Doppler ultrasound to assess slow flow in the small-diameter vasculature, such as the maternal spiral arteries and fetal villous arteries of the placenta and focal liver lesions. The sensitivity of PD in small vessel detection is limited by the low signal produced by slow flow and the noise associated with small vessels. The noise sources include electronic noise, stationary or slowly moving tissue clutter, reverberation clutter, and off-axis scattering from tissue, among others. In order to provide more sensitive detection of slow flow in small diameter vessels, a coherent flow imaging technique, termed coherent flow PD (CFPD), is characterized and evaluated with simulation, flow phantom experiment studies, and an in vivo animal small vessel detection study. CFPD imaging was introduced as a technique to detect slow blood flow. It has been demonstrated to detect slow flow below the detection threshold of conventional PD imaging using identical pulse sequences and filter parameters. In this paper, we compare CFPD with PD in the detection of blood flow in small-diameter vessels. The results from the study suggest that CFPD is able to provide a 7.5-12.5-dB increase in the signal-to-noise ratio (SNR) over PD images for the same physiological conditions and is less susceptible to reverberation clutter and thermal noise. Due to the increase in SNR, CFPD is able to detect small vessels in high channel noise cases, for which PD was unable to generate enough contrast to observe the vessel.

Short-lag spatial coherence imaging on matrix arrays, Part II: Phantom and in vivo experiments

In Part I of the paper, we demonstrated through simulation the potential of volumetric short-lag spatial coherence (SLSC) imaging to improve visualization of hypoechoic targets in three dimensions. Here, we demonstrate the application of volumetric SLSC imaging in phantom and in vivo experiments using a clinical 3-D ultrasound scanner and matrix array. Using a custom single-channel acquisition tool, we collected partially beamformed channel data from the fully sampled matrix array at high speeds and created matched B-mode and SLSC volumes of a vessel phantom and in vivo liver vasculature. 2-D and 3-D images rendered from the SLSC volumes display reduced clutter and improved visibility of the vessels when compared with their B-mode counterparts. We use concurrently acquired color Doppler volumes to confirm the presence of the vessels of interest and to define the regions inside the vessels used in contrast and contrast-to-noise ratio (CNR) calculations. SLSC volumes show higher CNR values than their matched B-mode volumes, while the contrast values appear to be similar between the two imaging methods.

A GPU-based real-time spatial coherence imaging system

Advanced ultrasonic beamforming techniques are often computationally intensive and difficult to implement in real-time. GPU computing has become a vital tool for software beamforming because of its massive parallel computing capabilities. However, GPU-based software beamforming has not yet been integrated into a real-time imaging system. We have recently introduced short-lag spatial coherence (SLSC) imaging as a coherence-based beamforming technique that is more robust to clutter than conventional B-mode imaging. The algorithm is computationally expensive, and has been limited to offline processing to date. By combining SLSC beamforming on the GPU with a Verasonics ultrasound scanner, we have realized a real-time side-by-side B-mode and SLSC imaging system capable of achieving up to 6 frames per second (fps). We demonstrate the systems real-time capabilities with phantom and in vivo scans, and briefly examine the relative performance of B-mode and SLSC imaging.

Development and evaluation of pulse sequences for freehand ARFI imaging

Acoustic Radiation Force Impulse (ARFI) imaging techniques can be used to non-invasively evaluate the relative stiffness of tissue; potentially aiding in the identification of stiff cancerous lesions or vulnerable soft lipid filled plaques in vasculature. In this work, we developed several pulse sequences that implement conventional B-mode interleaved with ARFI imaging for various frame rates and scan durations (ranging from 20 fps for 1 sec to 1 fps for 90 seconds). On-axis displacement estimates were calculated using a GPU processor for faster computation times. The clinical feasibility of the proposed pulse sequences was evaluated on the basis of transducer face heating, ISPTA, MI, and the observed spatial and temporal consistency of images acquired in vivo. Transducer face heating was <;2°C and FDA acoustic exposure limits were not exceeded. In phantoms, image degradation during a freehand swept scan is shown to be dependent upon the sonographer. In vivo multi-frame movie sequences showed consistent measurements of on-axis displacements across multiple frames and acquisitions. Overall data acquisition, processing, and display frame rates of 1.0 Hz were achieved, allowing for low frame rate sequences to provide feedback information during the swept scan.

In vivo demonstration of a real-time simultaneous B-mode/spatial coherence GPU-based beamformer

SLSC imaging is an advanced ultrasound beamforming technique that generates images of the coherence in the backscattered echo and has been shown to suppress image degradation due to clutter. However, the computational demand and the need for access to channel data has made real-time implementation difficult. GPU computing can significantly improve computational throughput, but has been used primarily to accelerate offline image processing. By implementing the SLSC algorithm on 3 GPUs and integrating it with a Verasonics ultrasound scanner, we have realized a real-time simultaneous B-mode and SLSC imaging system. We demonstrate the real-time imaging capabilities on an in vivo liver, gallbladder, and carotid artery and present snapshots of the on-screen display. The SLSC images were found to suppress clutter and showed a more homogeneous texture in tissue when compared to the B-mode images. Computation time and imaging frame rate were highly dependent on the imaging parameters. When imaging the liver, 46 transmit beams with a range of 10 cm were obtained at 6.7 fps for standard imaging and at 2.1 fps for harmonic imaging. A frame rate of 5.9 fps was achieved for carotid artery imaging with 129 transmit beams and a range of 3 cm. By restricting the field of view to 65 transmit beams and a range of 2 cm, a frame rate of 12.5 fps was achieved. Further increases in frame rates can be attained with additional system development and improvements to hardware.

Application of synthetic aperture focusing to short-lag spatial coherence

It has been demonstrated that short-lag spatial coherence (SLSC) ultrasound imaging can provide improved SNR and CNR compared to conventional B-mode images, especially in the presence of noise and clutter. Application of the van Cittert-Zernike theorem predicts that coherence between the ultrasound echoes received across an array is reduced significantly away from the transmit focal depth, leading to a limited axial depth of eld in SLSC images. Transmit focus throughout the eld of view can be achieved using synthetic aperture methods to combine multiple transmit events into a single nal image. We propose the application of these methods to create synthetically focused channel data to be used to create an SLSC image that will have an extended axial depth of eld comparable to B-mode images. Experimental results in a phantom and in vivo are presented and compared to both B-mode images and dynamic receive focused SLSC images, demonstrating improved SNR and CNR away from the transmit focus and an enlarged axial depth of eld.

Real-time high-framerate in vivo cardiac SLSC imaging with a GPU-based beamformer

Echocardiography is widely used to evaluate cardiovascular function in real-time; however, endocardial borders are often inadequately visualized due to clutter. Contrast agents, while effective, increase cost and examination time. SLSC imaging is a coherence-based beamforming technique that has demonstrated clutter reducing capabilities in a variety of applications. Previously, real-time SLSC imaging of a large field of view was achieved at up to 6 fps, inadequate for visualizing a quickly moving target such as the heart. In our previous work, we developed efficient modifications to the SLSC algorithm to obtain fast and precise measurements of spatial coherence. In this work, these techniques were implemented on a GPU-based software beamformer to develop a real-time SLSC imaging system suitable for echocardiography. The system was then used in a clinical study to image 15 patients with poor image quality, as determined by trained sonographers. Real-time display of side-by-side B-mode and SLSC images were obtained at frame rates ranging from 21 to 31 fps. B-mode and SLSC video clips were given in shuffled order to a cardiologist, who rated the visibility s of each of 12 LV segments as 0=invisible, 1=poorly visualized, or 2=well visualized. Scores of 0 and 1 indicated a need for contrast agent. A two-tailed paired t-test showed that SLSC imaging demonstrated a statistically significant increase in the number of visualized segments (s > 0, p = 0.0032) and in the number of segments that were well visualized (s = 2, p = 0.0061).