光声信号频谱分析

发布时间：2018-12-14 00:49

【摘要】：生物组织光声成像是一种新型的医学成像技术,它同时具有光学成像的强对比度和超声深层组织成像分辨率高的优势,能够对生物组织提供结构的、功能的和分子的成像。在光声成像中,激光照射生物组织,由于光声效应产生光声信号。信号通过一个宽带的换能器接收,小信号放大器放大,数据采集卡采样等步骤,得到了储存在计算机上的光声信号。目前,对生物组织光声信号的研究主要集中在时域分析。可是,时域信号受到各种外界因素的影响,比如激光光源的激发脉冲、换能器的频率响应、测量系统的传递函数等。这些因素导致时域分析得到的图像只能够定性地反映光吸收体的相对强弱。这样就存在一个问题,即对同样的一个样品,不同的仪器甚至相同的仪器不同的操作者得到的结果无法做到定量的比较。本文以定量化为目标,在频域上对光声信号作了深入的研究,并取得了以下研究成果：本文首先从理论上研究了随机分布的单一尺寸光吸收体的光声信号频谱特征。理论分析表明功率谱斜率受到激光照射函数、光吸收体的相关函数、指向性函数、换能器响应和估计参数的带宽等因素影响,并且对这些因素的影响作了定量的分析。然后,通过一个理想的实验证明了上述理论方程。实验表明：当通过校准的方式移除外界的影响因素之后,能够得到只和组织的微结构有关的频谱信息。最后,我们从校准信号中取出斜率参数,发现不同的系统得到的校准之后的斜率参数可以进行定量的比较。即我们从理论上和实验上证明了校准的光声频谱信号具有和操作者以及设备无关的特征。这个工作对以后做定量化测量及比较是非常有意义的。本文根据对光声信号频域理论分析,提出了用频谱拟合的方式去测量深埋在浑浊媒质中随机分布的光吸收体尺寸的方法。频谱理论指出,功率谱信号的低频段和光吸收体尺寸具有一一对应的关系。因此,可以从低频段提取参数信息,去定量地测量粒子尺寸。理想的样品实验证明了上述方法的有效性。我们准确地测量出三种微颗粒的尺寸信息。由于频谱方法可以使用低频段的光声信号来测量样品,所以该方法适用于较深处组织的定量测量。此外,本文以血细胞聚集这一现象为模型,对混合尺寸的随机分布微颗粒的光声信号进行了理论分析,给出一般性的频谱表达式。由于血细胞的聚集情况比较复杂,我们引入等效粒子尺寸的概念,定量的描述了混合粒子的整体特征。最后,实验得到的混合粒子的特征尺寸和理论吻合很好。因此,用混合粒子的频谱进行定量测量扩展了单一尺寸的应用模型,具有更实际的应用价值。
[Abstract]:Biological tissue photoacoustic imaging is a new type of medical imaging technology. It has the advantages of strong contrast of optical imaging and high resolution of ultrasonic deep tissue imaging, which can provide structural, functional and molecular imaging for biological tissue. In photoacoustic imaging, photoacoustic signals are produced by laser irradiation on biological tissues. The signal is received by a wideband transducer, amplified by a small signal amplifier and sampled by a data acquisition card. The photoacoustic signal stored on the computer is obtained. At present, the research of biological tissue photoacoustic signal is mainly focused on time domain analysis. However, the time domain signal is affected by various external factors, such as the excitation pulse of the laser source, the frequency response of the transducer, the transfer function of the measurement system, and so on. These factors can only reflect the relative strength of optical absorber qualitatively. The problem is that for the same sample, different instruments and even the same instrument, the results obtained by different operators can not be compared quantitatively. Aiming at quantification, the photoacoustic signal in frequency domain is studied deeply in this paper, and the following research results are obtained: firstly, the spectrum characteristics of photoacoustic signal with random distribution of single size photoabsorber are studied theoretically in this paper. Theoretical analysis shows that the slope of power spectrum is affected by the factors such as laser irradiation function, correlation function of optical absorber, directivity function, transducer response and the bandwidth of estimating parameters, and the influence of these factors is analyzed quantitatively. Then, the above theoretical equations are proved by an ideal experiment. The experimental results show that the spectrum information related only to the microstructure of the tissue can be obtained when the external factors are removed by means of calibration. Finally, we take the slope parameters from the calibration signal and find that the slope parameters after calibration can be quantitatively compared with each other. That is, we prove theoretically and experimentally that the calibrated photoacoustic spectrum signal has characteristics independent of operator and equipment. This work is very meaningful for quantitative measurement and comparison in the future. Based on the theoretical analysis of the frequency domain of photoacoustic signals, this paper presents a method of measuring the size of light absorbers with random distribution deeply buried in turbid media by means of spectrum fitting. The spectrum theory indicates that the low frequency band of the power spectrum signal has a one-to-one correspondence with the size of the optical absorber. Therefore, the parameter information can be extracted from the low frequency band to measure the particle size quantitatively. The experimental results show that the proposed method is effective. We accurately measured the size information of three kinds of microparticles. Because the spectrum method can be used to measure samples with low frequency band photoacoustic signals, this method is suitable for quantitative measurement of deeper tissues. In addition, taking the phenomenon of blood cell aggregation as a model, the photoacoustic signals of randomly distributed microparticles of mixed size are theoretically analyzed, and a general spectrum expression is given. Because the aggregation of blood cells is complicated, we introduce the concept of equivalent particle size and describe the global characteristics of mixed particles quantitatively. Finally, the characteristic size of the mixed particles is in good agreement with the theory. Therefore, the application model of single size is extended by using the spectrum of mixed particles, which has more practical application value.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：Q6-33;TN911.6

【相似文献】