REFoCUS (Retrospective Encoding For Conventional Ultrasound Sequences) offers great flexibility by enabling synthetic aperture beamforming from conventional ultrasound sequences. This flexibility is beneficial for many aspects in medical ultrasound beamforming, including e.g. combination of different transmit waves, distributed sound speed estimation and common-midpoint gathers. REFoCUS beamforming also has image quality comparable to state-of-art methods such as Retrospective Transmit Beamforming (RTB). However, the previously published implementations of REFoCUS do not address clutter from sidelobes and grating lobes present in the data before the recovery. This reduces image quality due to potentially strong sidelobes and grating lobes, particularly when using REFoCUS in combination with micro-beamforming and matrix array probes. Recordings from micro-beamforming probes may thus not be compliant with the existing REFoCUS methods. We propose to solve the sidelobes and grating lobe issues by introducing a reformulation of REFoCUS that performs multistatic data recovery and beamforming in the time domain, allowing spatial weighting to remove clutter and noise. Spatial weighting is based on common beamforming principles and incorporates element directivity, dynamic F-number, beam geometry weighting, and grating lobe suppression. We also discuss how aperture sampling affects beamforming with REFoCUS. Spatially Weighted REFoCUS (SWR) and critical sampling of the transmit aperture show suppression of receive grating lobes in an in vivo setting with two different micro-beamforming matrix-array probes, leading to an increase in gCNR contrast from 0.44 to 0.96 in a fetal image and from 0.39 to 0.89 in a cardiac image.

Universal REFoCUS beamforming with spatial weighting

Optimized Apodizations with Training Simulations (OATS): Learned depth-dependent apodizations via differentiable beamforming for reduced operator tuning


 Medical ultrasound instruments have been available for clinical users for roughly the past 60 years. During this time the field has received deep attention from researchers in academia and industry. There has also been a close coupling between ultrasound scanners and microelectronics which has significantly aided in the process. In this invited talk this history will be summarized with the goal of defining the starting point of future technical developments of the field.

Possible futures for medical ultrasound technology

We present vbeam: a new open-source beamforming library that provides fast parallel beamforming computation on the CPU and GPU for modern imaging setups. It is auto-differentiable which allows for incorporating image optimization and machine learning (ML) into the beamformer, and vice versa. The library can be used from either Python or MATLAB and aims to be a beamforming research tool which can be used standalone or integrated with tools, such as USTB, Field II and MUST.The vbeam library is implemented on top of Google’s JAX ML library but is extendable to other backends, such as Tensorflow and PyTorch. This backend handles all device computation, and allows for taking the gradient of the beamformer. The vbeam library provides a functional and extendable API for defining beamformers, currently supporting both conventional virtual source wavefront models and a time-domain implementation of REFoCUS.As an example, we demonstrate that vbeam can be used with gradient descent to optimize an adaptive apodization function that minimize the variance of the delayed signal across receiving elements.Beamformers defined in vbeam processes in the order of 10 billion points per second on a Nvidia A100 GPU, depending on the imaging setup. Our time-domain implementation of REFoCUS beamforms an image in 850 milliseconds on average. When additionally modeling a 150 × 150 heterogeneous speed-of-sound map it takes 1920 milliseconds on average. A single gradient descent step for optimizing the adaptive apodization function takes 120 milliseconds on average.

vbeam: a Fast and Differentiable Beamformer for Optimizing Ultrasound Imaging

Synthetic transmit focusing are techniques that synthesizes transmit focus retrospetively in the recorded channel data from multiple transmissions. Synthetic transmit focusing has a large impact on the estimated coherence in medical ultrasound imaging, since the coherence is dependent on the type of trans-mitted wave and, for focused transmissions, the location of the focus. Retrospective Transmit Beamforming (RTB) is currently the state-of-the-art for focused transmit b-mode imaging in clinical scanners. An alternative approach for synthetic transmit focusing, REFoCUS (Retrospective Encoding For Conventional Ultrasound Sequences) can reconstruct the full synthetic transmit aperture imaging dataset. We compare the resulting b-mode images and coherence obtained from RTB, REFoCUS and conventional scanline beamforming and demonstrate that REFoCUS and RTB have comparable results. However, REFoCUS has slight improvements in terms of contrast, resolution and coherence compared to RTB as well as the main advantage of REFoCUS; that it is less dependent on the fine tuning of delay models and apodization needed for RTB. Thus, REFoCUS is a powerful tool for software beamforming.

Coherence from REFoCUS compared to Retrospective Transmit Beamforming

In beamforming, retrospective change in sound speed and recalculation of focusing delays is attractive both for improving image quality and for using it in an iterative image quality optimization process. Modifying the speed of sound retrospectively for focused transmits is challenging because the transmit focus position is a function of sound speed error. The virtual source model is a common way to calculate the transmit focusing delays where using the correct transmit focus position is imperative. In this paper, we provide the methods necessary to perform a retrospective sound-speed correction by compensating the receive grid and by calculating the effective transmit focus needed to perform proper synthetic transmit focusing. To evaluate the efficacy of our method, we simulate wave propagation and measure the resolution of in vitro images using both phased and curvilinear arrays. The results of the suggested virtual source estimation method match the simulated wave propagation for multiple F-numbers and both positive and negative sound speed errors. We compare beamformed images using correct/incorrect sound speeds and correct/incorrect virtual source positions. The results demonstrate that the Corrected Virtual Source (CVS) method generates artifact-free images with superior quality compared to images with incorrect sound speed. Furthermore, the image beamformed with the correct sound speed, but incorrect virtual source position, exhibits image artifacts and inferior focusing quality compared to the CVS image.

Sound Speed and Virtual Source Correction in Synthetic Transmit Focusing

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Anders Vrålstad

Author Tools

Chat about Author

Papers

vbeam: a Fast and Differentiable Beamformer for Optimizing Ultrasound Imaging

Coherence from REFoCUS compared to Retrospective Transmit Beamforming

Universal REFoCUS beamforming with spatial weighting

Sound Speed and Virtual Source Correction in Synthetic Transmit Focusing