基于虚拟点探测的压缩感知快速活体光声成像方法研究

项目名称： 基于虚拟点探测的压缩感知快速活体光声成像方法研究

项目编号： No.61308116

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 无线电电子学、电信技术

项目作者： 孟静

作者单位： 曲阜师范大学

项目金额： 27万元

中文摘要： 光声成像兼具纯光学成像方法的对比度、光谱特异性和超声对深组织成像的高空间分辨率，在生物医学成像领域具有重要应用价值。阵列式光声计算层析成像具备快速临床成像潜力，然而，大规模超声阵列数据采集量大，对数据采集、传输和重建都提出了很高的要求。为此，本课题将研究稀疏信号采样模式下的压缩感知快速光声成像方法。首先，研究基于聚焦超声探头的虚拟点探测-稀疏信号采集方案，提供大接收角度和高灵敏度的信号探测；其次，研究并行压缩感知光声成像方法，建立相应的并行计算框架和数学模型，致力于实现满足临床应用需求的光声图像采集、重建和显示帧率；最后，开展小动物活体成像实验，拟通过小鼠脑部灌注成像评估系统的快速成像能力，小鼠断层和三维成像验证系统成像质量。课题的实施，将为快速、低成本光声成像系统开发提供理论依据，推动光声成像技术在生物医学研究及疾病诊断与治疗评估中的深入应用。

中文关键词： 光声成像；虚拟点探测；压缩感知；GPU并行计算；图像重建

英文摘要： Photoacoustic imaging combines the physics of optical and ultrasonic imaging to provide both the high optical contrast and spectroscopic specificity and the high spatial resolution of ultrasound for deep tissue imaging. So far, it has been demonstrated to have broad applications in biomedicine. Photoacoustic computed tomography with ultrasonic array has the potential of high-speed imaging in the clinic, however, the acquired data amount is huge, and thus the data acquisition, transmission and reconstruction are all challenged. In this proposal, high-speed photoacoustic tomography under sparse sampling will be developed based on compressed sensing technique. First, to achieve high sensitivity and large acceptance angle in the measurement, virtual point detection scheme of sparse signals with focused transducers of large numerical apertures will be studied. Second, to meet the image frame rates required by clinical use, parallel compressed-sensing photoacoustic imaging methods will be developed，and the corresponding parallel computation framewrok and mathematical model will be established. Finally, in vivo small animal experiments, in particular, rat cerebral perfusion imaging will be carried out to evaluate the system's imaging speed, while small-animal whole body imaging will be used to assess the quality of rec

英文关键词： photoacoustic imaging；virtual-detector；compressed sensing；GPU parallel computation；image reconstruction