超分辨超声成像的微泡声学特性与识别方法研究

High-resolution imaging of cerebral blood flow can provide important data for brain science. At present, conventional blood flow detection methods are not able to meet the needs of high-resolution cerebral blood flow imaging in the whole brain. The newly proposed super-resolution ultrasound technology uses plane wave ultrasound in combination with micro-bubbles to achieve a high resolution blood flow imaging by breaking ultrasound diffraction limit. However, the frame rate of current method is low (more than 30 second for one image), and the acoustic characteristics of different ultrasound microbubbles have a great influence on the imaging. This project is planned to study the acoustic echo characteristics of different ultrasonic microbubbles with different sizes and concentrations under different ultrasonic frequencies. Previous works on high-resolution ultrasound imaging and characterization of microbubbles have paved a good way for this study. The acoustic properties of ultrasonic microbubbles and background tissue noises are analyzed. The acquired parameters are used to compensate for the optimization of super-resolution imaging microbubble recognition algorithm to improve the accuracy of recognition. A new super-resolution imaging platform will be built up with parallel computing architecture to increase the imaging frame rate. In vivo animal study will be carried out. The goal of this study is to achieve more accurate super-resolution ultrasound imaging methods that accurately characterize the blood flow distribution at the whole brain level in small animals. The implementation of this project is expected to provide new blood flow imaging methods and tools for brain science, and can significantly expand the related applications of ultrasound imaging.

脑血流的高分辨率成像可为脑科学研究提供重要的功能性数据。目前常规脑血流检测手段都无法兼顾高分辨率和成像深度需要。最新提出的超分辨超声成像利用平面波超声与微泡相结合实现了突破超声衍射极限的血流成像。然而目前成像帧频较低(一帧图像需大于30秒)，且不同超声微泡的物理参数包括尺寸和浓度对超分辨成像的精度有较大影响。本项目基于前期高分辨率超声成像、流体速度场成像和微泡制备与表征的研究基础，拟研究不同超声微泡尺寸、浓度在不同超声频率作用下的声学回波特性；分析超声微泡声学特性与背景组织噪声的差异；利用获取的关系参数补偿优化超分辨成像微泡识别算法，提高识别的准确性；基于搭建的并行计算架构的超分辨成像系统，开展小动物成像实验研究。目标实现精准的超分辨超声成像方法用以表征小动物全脑层面的血流分布。本项目的实施有望为脑科学研究提供新的血流成像方法及工具，并可显著拓展超声成像的应用领域。

脑血流的高分辨率成像可为脑科学研究提供重要的功能性数据。目前常规脑血流检测手段都无法兼顾高分辨率和成像深度需要。最新提出的超分辨超声成像利用平面波超声与微泡相结合实现了突破超声衍射极限的血流成像。然而目前成像帧频较低(一帧图像需大于30秒)，且不同超声微泡的物理参数包括尺寸和浓度对超分辨成像的精度有较大影响。本项目基于前期高分辨率超声成像、流体速度场成像和微泡制备与表征的研究基础，研究了超声微泡在超声作用下的声学回波特性，分析了超声微泡声学基波和谐波特性；设计开发了超分辨率超声成像算法，并利用获取的微泡声学特性优化超分辨成像微泡识别算法；搭建了并行计算架构的超分辨超声成像系统，开展了仿体内的超声成像研究，并最终开展了小动物活体脑部组织在体超分辨超声成像实验研究。项目实现了超分辨超声成像方法，并用以表征小动物脑部血流分布。本项目研究的高分辨率超声成像方法可以为脑科学研究提供新的血流成像方法及工具，并拓展了超声成像的应用领域。