The Spectral Analysis Of Photoacoustic Signal

Photoacoustic tomography (PAT) is an emerging technique, which inherits the merits of the high contrast of optical imaging and the high resolution of ultrasonic imaging. It can provide the anatomical, functional and molecular imaging of biological tissues. In PAT, the laser irradiates on the tissues. The generated photoacoustic signals are detected by the ultrasound transducer with a broad bandwidth, amplified and recorded by a data acquisition card. The digitalized signals are stored in computer for the further analysis.The analysis of photoacoustic signal is mainly focused on the time domain at present. But the time domain signals can be affected by many external factors such as the pulse laser, the response of transducer and the transfer function of the measurement system, etc. The image reconstructed from the time domain can only reflect the contrast of the optical absorbers qualitatively. Therefore, it is difficult to quantify the measurement results from different systems, even the same system operated by different users. This dissertation focuses on the quantification of photoacoustic signals and the major work is given as follows:We theoretically deduce that photoacoustic spectral characteristics depend on the particle size and extrinsic factor. These factors are quantitatively studied in this dissertation. And an ideal experiment has demonstrated the spectral formula. The results represent that the spectrum can only related with the microstructure after calibration. At last, the parameter extracted from the calibrated spectrum obtained by different system can be quantitatively compared. We have experimentally and theoretically demonstrated that the calibrated spectrum has the user-independent and system-independent features. This work would be meaningful for the quantitative measurement and comparison of the tissues.According to the analysis of the photoacoustic spectrum, we propose a spectral matching method to quantitatively measure the stochastic microstructure in turbid and deep media. The spectral theory indicates that the low domain of the spectrum can be one-to-one correspondence with the size of the optical absorbers. So the parameter can be extracted from the low frequency domain to measure the size of the absorber. We have accurately measured three kinds of size of microstructures experimentally. Since the spectral method can use the low working frequency of photoacoustic signals, it can be used for the quantitative measurement in deep tissues.In addition, basing on the aggregation of red blood cells, we theoretically deduce the power spectrum function of the photoacoustic signal from the tissue with non-uniform aggregations. In order to describe the overall features, the equivalent characteristic size is proposed to quantify the microstructure feature of the granular aggregations. And results with test phantoms validate the theory and the feasibility of the measurement. So the spectral theory of mixed particles would be more universal for medical applications.