【作者】 李帅；

【导师】 李娇；

【作者基本信息】 天津大学 ， 生物医学工程， 2022， 硕士

【摘要】 光声成像具有高光学对比度、高成像深度和高分辨率等特点,是近年来最有前景的生物医学成像技术之一。传统的光声成像使用超声换能器接收光声信号,这种接触式的探测方法给系统设计、成像性能和应用带来一定局限。近年来,一系列非接触超声探测方法凭借其高带宽、高灵敏度和无耦合剂的特点,在继承光声成像优势的基础上,能提升探测性能、拓宽应用范围。其中,基于弹光效应的非相干光声遥感探测(Photoacoustic Remote Sensing,PARS)具有完全非接触、鲁棒性好、灵敏度高等特点。PARS的成像速度、成像视野和成像质量是其向临床转化的关键指标,目前仍有巨大提升空间。因此,本文在探索物理机理的基础上,搭建了PARS硬件成像系统,提出了新型的控制策略和高效的重建算法,并开发了集成控制界面实现可实时反馈的快速大范围高分辨PARS成像。机理研究方面。针对PARS非接触探测中的“光声–弹光”物理过程,结合时域有限差分方法,本文提出了基于透明源的嵌套式时域有限差分数值计算方法。该方法能够同时考虑声压传输和折射率调制,实现探测光的反射光强精准仿真。硬件系统与控制策略方面。本文基于“光声-弹光”原理,搭建了一套PARS系统,奠定了成像的硬件基础。为在高脉冲重复频率下提升成像速度,提出了基于实时数据流的采集策略,其能实现PARS信号采集、传输和处理高效并行化,为后续成像提供高速原始数据流。为了扩大系统的成像视野,在PARS系统和实时数据流的基础上,本文提出了一种光机联合扫描策略,通过坐标系变换将振镜高速扫描和平移台大范围扫描精确配准,以实现大范围的高分辨快速成像。算法研究与集成控制方面。图像重建算法是提高成像质量的关键手段之一,针对发展的PARS成像系统和控制策略,本文发展了一种散点快速网格插值方法,能够在远小于扫描时间的条件下重建得到高质量的PARS图像。综合PARS硬件系统、控制策略和重建算法,开发了PARS-Studio集成控制界面,实现可实时反馈的PARS在线处理与成像。最后,本文对所发展的PARS成像系统进行了一系列实验研究。仿体实验证明该系统具备良好的成像分辨率、成像速度、成像视野和定量能力。小鼠和斑马鱼活体实验进一步验证了所发展的控制策略和重建算法能够推动PARS成像技术在预临床研究的发展进程,可为生物医学和材料检测领域提供有力成像工具。更多还原

【Abstract】 Photoacoustic imaging is one of the most promising biomedical imaging techniques in recent years due to its characteristics of high optical contrast,high imaging depth and high resolution.Traditional photoacoustic imaging uses ultrasonic transducers to receive photoacoustic signals.This contact detection method brings limitations to system design,imaging performance and application.In recent years,a series of non-contact ultrasonic detection methods have been shown to obtain high bandwidth and high sensitivity without coupling agents.Inheriting the advantages of photoacoustic imaging,those non-contact methods can improve the detection performance and broaden the application range.Among them,photoacoustic remote sensing(PARS)based on the elastic-optical effect has features of complete non-contact,good robustness and high sensitivity.However,the imaging speed,field of view and imaging quality of PARS still need to be improved for the further pre-clinical and clinical transformation.Therefore,this paper builds a PARS hardware imaging system based on the physical mechanism,and proposes new control strategies and the corresponding efficient reconstruction algorithm.An integrated control interface is also developed to achieve real-time feedback,rapid,large-field and high-resolution PARS imaging.For mechanism research,combined with the finite difference time domain(FDTD)method,a nested FDTD numerical calculation method based on transparent source is proposed for the physical process of "photoacoustic-elastic-optical" in PARS detection.This method can take pressure transmission and refractive index modulation into account,and realize accurate simulation of reflected light intensity.For hardware system,based on the principle of "photoacoustic – elastic-optical",a PARS system is built and lays the hardware foundation of imaging.For control strategy,a real-time data streaming acquisition strategy is proposed to improve the imaging speed at high pulse repetition rate.This control strategy can realize efficient parallelization of PARS signal acquisition,data transmission and processing,providing high-speed original data stream for PARS imaging.In order to enlarge the field of view,an opto-mechanical joint scanning strategy is proposed,based on the developed PARS system and real-time data stream.This strategy can accurately register coordinates between the fast scanning of the galvanometer and the large-range scanning of the translation stages,for achieving high resolution and fast PARS imaging with a large field of view.Image reconstruction algorithm is one of the key means to improve the image quality.In this paper,a fast scatter grid interpolation method is developed for the proposed PARS imaging system and control strategies,which can reconstruct high quality PARS images using far less scanning time.By integrating PARS hardware system,control strategy and reconstruction algorithm,an integrated control interface(PARS-Studio)is developed to realize real-time feedback PARS with online processing and imaging.Finally,a series of experiments are carried out on the proposed PARS imaging system.The results of phantom experiments show that the system has good image performance of spatial resolution,imaging speed,field of view and quantitative ability.In vivo experiments on the mice and the zebrafish further verify that the proposed control strategies and reconstruction algorithm can promote the development of PARS imaging technology in pre-clinical research,providing a powerful imaging tool for biomedical applications and the material detection.更多还原