Accelerating compute intensive non-real-time beam-forming algorithms in
ultrasound imaging using deep learning architectures has been gaining momentum
in the recent past. Nonetheless, the complexity of the state-of-the-art deep
learning techniques poses challenges for deployment on resource-constrained
edge devices. In this work, we propose a novel vision transformer based tiny
beamformer (Tiny-VBF), which works on the raw radio-frequency channel data
acquired through single-angle plane wave insonification. The output of our
Tiny-VBF provides fast envelope detection requiring very low frame rate, i.e.
0.34 GOPs/Frame for a frame size of 368 x 128 in comparison to the
state-of-the-art deep learning models. It also exhibited an 8% increase in
contrast and gains of 5% and 33% in axial and lateral resolution respectively
when compared to Tiny-CNN on in-vitro dataset. Additionally, our model showed a
4.2% increase in contrast and gains of 4% and 20% in axial and lateral
resolution respectively when compared against conventional Delay-and-Sum (DAS)
beamformer. We further propose an accelerator architecture and implement our
Tiny-VBF model on a Zynq UltraScale+ MPSoC ZCU104 FPGA using a hybrid
quantization scheme with 50% less resource consumption compared to the
floating-point implementation, while preserving the image quality.

利用深度学习架构来加速超声成像中的计算密集型非实时波束形成算法近来愈加受到关注。然而，现有最先进的深度学习技术的复杂性对于资源有限的边缘设备的部署提出了挑战。本文提出了一种基于视觉转换器的微型波束形成器 (Tiny-VBF)，它使用通过单角平面波入射获得的原始射频通道数据。我们的 Tiny-VBF 的输出提供了快速的包络检测，仅需要非常低的帧率，即对于 368 x 128 的帧大小，每帧仅需要 0.34 GOPs / 帧，相比于最先进的深度学习模型。与 Tiny-CNN 在体外数据集上相比，它还显示出了 8% 的对比度增加、5% 和 33% 的轴向和横向分辨率的增益。另外，当与传统的延迟和求和 (DAS) 波束形成器进行比较时，我们的模型显示出了 4.2% 的对比度增加和 4% 和 20% 的轴向和横向分辨率的增益。我们还提出了一种加速器架构，并使用混合量化方案在 Zynq UltraScale+ MPSoC ZCU104 FPGA 上实现了我们的 Tiny-VBF 模型，其资源消耗比浮点实现少 50%，同时保持了图像质量。