Efficient FPGA Firmware Coding for High Frame Rate Ultrasound Imaging

这个博士建议适合几个较大的跨学科项目，旨在生产诊断和治疗超声成像工具。整个项目的目标是为可配置的高速超声研究工具（称为超声阵列研究平台（UARP））开发FPGA固件代码。UARP项目是一个具有三种实现方式的超声平台：用于诊断成像的高速128通道系统，用于工业应用的16通道离散换能器系统，以及高强度聚焦超声（HIFU）系统，称为HIFU阵列研究平台（HIFUARP）。所有UARP平台都集成了一系列高速ADC和FPGA，可以执行各种成像和治疗模式。研究强调需要同时使用成像和HIFUARP，用于超声引导的HIFU治疗。该项目将首先开发一种方法来控制成像和治疗系统作为一个，精确的定时，使先进的多模式序列相结合的成像和治疗超声的发展。这项研究将集中在设计和实现问题的一个高效的信号/图像处理子系统的非常高的帧速率超声成像系统。这项工作提出了调查各种自适应波束形成算法，可以提供运行时的可重构性，根据不断变化的应用。自适应平面波波束形成等技术将在可重构硬件上实现，具有运行时的适应性。为了解决超声成像中伪影去除的挑战，将研究具有运行时可重构性的附加特征的自适应图像处理算法。这项工作将探索以前在自适应图像处理算法领域所做的努力，并根据应用的性质训练可重构图像处理系统以适应最佳技术。该项目的成果将进一步开发用于临床原型的成像和治疗套件。

This PhD proposal fits within several larger interdisciplinary projects aimed at producing a diagnostic and therapeutic ultrasound imaging tools. The objective of the overall project is to develop FPGA firmware code for a configurable high speed ultrasonic research tool known as the Ultrasound Array Research Platform (UARP). The UARP project is an ultrasound platform with three implementations; a high speed, 128 channel system for diagnostic imaging, 16 channel discrete transducer system for industrial applications, and high intensity focused ultrasound (HIFU) system, known as the HIFU Array Research Platform (HIFUARP). All the UARP platforms incorporate an array of high speed ADCs and FPGAs and can perform variety of imaging and therapeutic modalities. Research highlights a need to use both the imaging and HIFUARP at the same time, for ultrasound guided HIFU therapy. The project will first develop a method to control both the imaging and therapy systems as one, with precise timing to allow the development of advanced multi-mode sequences combining imaging and therapeutic ultrasound.This research will then focus on design and implementation issues of an efficient signal/image processing sub-system for a very high frame rate ultrasonic imaging system. This work proposes to investigate various adaptive beamforming algorithms which may provide run-time reconfigurability in accordance to changing applications. Techniques like adaptive plane wave beamforming will be investigated for implementation in reconfigurable hardware with feature of run-time adaptability. To tackle the challenges of artefact removal in ultrasound imaging, adaptive image processing algorithms will be investigated with additional feature of run-time reconfigurability. The work will explore previous efforts made in the field of adaptive image processing algorithms and train the reconfigurable image processing system to adapt to the best technique depending upon nature of the application.The outcomes from the project will further develop the imaging and therapy suite for a clinical prototype.