# Deep Learning-based Universal Beamformer for Ultrasound Imaging

## TL;DR

This paper introduces a novel deep learning-based adaptive beamformer for ultrasound imaging that outperforms traditional methods in various data acquisition scenarios, providing higher quality images with improved contrast and structural similarity.

## Contribution

The paper presents the first data-driven deep neural network beamformer that is robust across different detector configurations and subsampling rates in ultrasound imaging.

## Key findings

- Significant improvement in contrast-to-noise ratio.
- Enhanced structural similarity of images.
- Effective for both focused and plane wave ultrasound imaging.

## Abstract

In ultrasound (US) imaging, individual channel RF measurements are back-propagated and accumulated to form an image after applying specific delays. While this time reversal is usually implemented using a hardware- or software-based delay-and-sum (DAS) beamformer, the performance of DAS decreases rapidly in situations where data acquisition is not ideal. Herein, for the first time, we demonstrate that a single data-driven adaptive beamformer designed as a deep neural network can generate high quality images robustly for various detector channel configurations and subsampling rates. The proposed deep beamformer is evaluated for two distinct acquisition schemes: focused ultrasound imaging and planewave imaging. Experimental results showed that the proposed deep beamformer exhibit significant performance gain for both focused and planar imaging schemes, in terms of contrast-to-noise ratio and structural similarity.