Research Of High-Speed And Multi-Channel Photoacoustic Tomography Imaging

Photoacoustic imaging is emerging as a novel biomedical imaging modality in the past two decades.It is a hybrid imaging modality using optical excitation,which combines the advantages of optical imaging with ultrasound imaging.As comparing with optical imaging,it can provide deeper imaging depth and higher spatial resolution.On the other hand,comparing with ultrasound imaging,owing to its reflecting specific absorption coefficient distribution of the tissue,it has higher optical contrast.All of these advantages make photoacoustic imaging widely used in clinical medicine,preclinical research and basic biological research.In recent years,photoacoustic imaging technology has made great advances,and photoacoustic imaging system is gradually developing towards real-time,multi-dimensional,so the laboratory project group upgrade the previous single-channel and single-element photoacoustic tomography imaging system and build a high-speed and multi-channel system.In this dissertation research of writing the LabVIEW acquisition control block diagram,implementation of high precision and fast image reconstruction algorithm and related imaging experimental verification are carried out.The main tasks in this paper are included three parts as follows:Firstly,the advantages and disadvantages of three kinds of photoacoustic image reconstruction algorithms,such as filtered back-projection,time reversal and model-based were analyzed.Combining on this system,a fast and high accuracy model-based algorithm with a novel and improved regularization method combining the L1 norm penalty of the reconstructed image after discrete wavelet and cosine transform(Discrete Cosine-Wavelet Transform Regularization,DCWT)and total variation(Total Variation Regularization,TV)was implemented,named DCWT-TV regularization,which could not only effectively reduce the strip artifacts produced by the traditional anti-projection reconstruction algorithm based on analytic solution,but also could effectively restore the boundary information and fine structure of the biological tissue in the image.Finally,the numerical simulation analysis was compared with the model-based Tikhonov regularization algorithm.Then using the LabVIEW programming language to write the block diagram of the acquisition and control program of the system and the main interface for user interaction.Simulated sample imaging experiments were carried out,the measurement of photoacoustic signal and the spatial resolution of system were tested.Finally,this system and reconstruction algorithm were validated by self-made sample imitation imaging experiment and healthy small nude mice and tumor bearing mice imaging experiment.The photoacoustic signal acquisition and storage process was only 1ms,and the image reconstruction time could be done within 40 s,this fully meet the requirements of rapid photoacoustic tomography imaging.