10 June 2024 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
A new approach for vascular super resolution imaging using the erythrocytes as targets (SURE imaging) is described and investigated. SURE imaging does not require fragile contrast agent bubbles, making it possible to use the maximum allowable mechanical index for ultrasound scanning for an increased penetration depth. A synthetic aperture ultrasound sequence was employed with 12 virtual sources using a 10 MHz GE L8-18i-D linear array hockey stick probe. The axial resolution was 1.20λ,(185.0μm) and the lateral resolution was 1.50λ,(231.3μm). Field IIpro simulations were conducted on 12.5 μm radius vessel pairs with varying separations. A vessel pair with a separation of 70 μm could be resolved, indicating a SURE image resolution below half a wavelength. A Verasonics research scanner was used for the in vivo experiments to scan the kidneys of Sprague-Dawley rats for up to 46 s to visualize their microvasculature by processing from 0.1 up to 45 s of data for SURE imaging, and for 46.8 s for super resolution (SR) imaging with a SonoVue contrast agent. Afterward, the renal vasculature was filled with the ex vivo micro-CT contrast agent Microfil, excised, and scanned in a micro-CT scanner at both a 22.6 μm voxel size for 11 hours, and for 20 hours in a 5 μm voxel size for validating the SURE images. Comparing the SURE and micro-CT images revealed that vessels with a diameter of 28 μm, five times smaller than the ultrasound wavelength, could be detected, and the dense grid of microvessels in the full kidney was shown for scan times between 1 to 10 s. The vessel structure in the cortex was also similar for the SURE and SR images. Fourier ring correlation indicated a resolution capability of 29 μm. SURE images are acquired in seconds rather than minutes without any patient preparation or contrast injection, making the method translatable to clinical use.
10 June 2024 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Super resolution ultrasound imaging using the erythrocytes (SURE) has recently been introduced. The method uses erythrocytes as targets instead of fragile microbubbles (MBs). The abundance of erythrocyte scatterers makes it possible to acquire SURE data in just a few seconds compared to several minutes in ultrasound localization microscopy (ULM) using MBs. A high number of scatterers can reduce the acquisition time, however, the tracking of uncorrelated and high-density scatterers is quite challenging. This paper hypothesizes that it is possible to detect and track erythrocytes as targets to obtain vascular flow images. A SURE tracking pipeline is used with modules for beamforming, recursive synthetic aperture imaging, motion estimation, echo canceling, peak detection, and recursive nearest neighbor tracker. The SURE tracking pipeline is capable of distinguishing the flow direction and separating tubes of a simulated Field II phantom with 125 to 25 μm wall-to-wall tube distances, as well as a 3D-printed hydrogel micro-flow phantom with 100 to 60 μm wall-to-wall channel distances. The comparison of an in-vivo SURE scan of a Sprague-Dawley rat kidney with ULM and micro-CT scans with voxel sizes of 26.5μm and 5μm demonstrated consistent findings. A microvascular structure composed of 16 vessels exhibited similarities across all imaging modalities. The flow direction and velocity profiles in the SURE scan were found to be concordant with those from ULM.
28 April 2023 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
This work presents a beamforming procedure that significantly reduces the number of operations when performing volumetric synthetic aperture imaging with row-column addressed arrays (RCAs). The proposed beamformer uses that the image values along the elevation direction of the low-resolution volume (LRV) are approximately constant. It is thus hypothesized that the entire LRV could be reconstructed from a single 2-D cross-section of the LRV. The presented method contains two stages; The first stage beamforms, for each emission, a cross-section using the conventional RCA beamformer. The second stage extrapolates the rest of the image points in the volume from the 2-D cross-sections. Assuming the image volume is covered by 3-D grid coordinates with a size of Nw × Nw × Nz , i.e., Nw samples along the x - and y -axis and Nz samples along the z -axis, the proposed beamformer reduces the number of mathematical operations by a factor of approximately N Nw /( NS + Nw ). Here S is the ratio between the first and second stage axial sampling rate, and N is the receiving aperture’s number of channels. Beamforming a 128 × 128 × 1024 volume from data acquired with N = 128 receiving channel can thus be achieved with 25.6 times fewer operations, when S = 4. A 9.23 times increase in the beamforming rate for a 100 × 100 × 200 volume with S = 2 was demonstrated on complex data from a 128+128 Vermon RCA probe. Real-time volumetric beamformation can, with this increase, be performed with a pulse repetition frequency of up to 1804.80 Hz. The proposed and conventional beamformer’s output was visually indistinguishable and, the full width at half maximum (FWHM) and full width at tenth maximum (FWTM) were at most 1.19% larger with the proposed approach. The proposed beamformer can thus perform volumetric imaging significantly faster than the current approach, with a negligible difference in image quality.
2 December 2024 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
An implementation of volumetric beamforming for row-column addressed arrays (RCAs) is proposed, with optimizations for Graphics Processing Units (GPUs). It is hypothesized that entire volumes can imaged in real time by a consumer-class GPU at an emission rate ≥12 kHz. A separable beamforming algorithm was used to reduce the number of calculations with a negligible impact on the image quality. Here, a single image was beamformed for each emission and then extrapolated to reproduce the volume, which resulted in 65 times fewer calculations per volume. Reusing computations and samples among adjacent pixels and frames reduced the amount of overhead and load instructions, increasing performance. A GPU beamformer, written in CUDA C++, was modified to implement the dual-stage imaging with optimizations. In-vivo rat kidney data was acquired using a 6 MHz Vermon 128+128 RCA probe and a Verasonics Vantage 256 scanner. The acquisition used 96 defocused emissions at a 12 kHz rate for a volume acquisition rate of 125 Hz. Processing time, including all pre-processing, was measured for an NVIDIA GeForce RTX 4090 GPU, and the resulting beamforming rate was 1440 volumes per second, greatly exceeding the real-time rate. Based on the GPU’s floating-point throughput, this corresponds to 22% of the theoretically achievable rate. High efficiency was also shown for an RTX 2080 Ti and RTX 3090, both achieving real-time imaging. This shows that 3D imaging can be performed in real time with a setup similar to 2D imaging: Using a single graphics card, one scanner, and 128 transmit/receive channels.
13 April 2026 in IEEE Transactions on Ultrasonics
Super-resolution ultrasound imaging using the erythrocytes (SURE) visualizes microvasculature in a few seconds without contrast agents. However, it is computationally intensive because it uses processing steps similar to those in ultrasound localization microscopy (ULM). The hypothesis is that a graphics processing unit (GPU) can process SURE in real time, allowing the resulting images to be displayed immediately at the patient’s bedside. The processing included beamforming, motion correction, tissue signal removal through singular value decomposition (SVD)-based filtering, and peak detection. All calculations were performed on the GPU, where batches of 32 frames were processed in parallel to improve hardware utilization. The image quality was compared with that of the original SURE implementation for in vivo rat kidney data acquired at 417 frames/s, and the processing rate was measured on an NVIDIA GeForce RTX 4090 GPU. The processing rate was 517 images/s, demonstrating real-time SURE imaging with GPU acceleration. Furthermore, the image quality was improved, with more blood vessels being clearly visualized. In conclusion, SURE imaging can be performed in real time, similar to power Doppler, enabling non-invasive super-resolution ultrasound imaging at the bedside.