The Research Of Denoising And Removing Artifact In Photoacoustic Imaging

Photoacoustic Imaging (PAI) is a novel imaging mode based on the photoacousticeffect. Since the lesions structures and the healthy parts are greatly different from theabsorb degree of laser in the biological tissue, PAI technology can provide a basicjudgment of the disease for the task of it is to rebuild the distribution of the absorbparts. PAI technology has both the advantage of high resolution from optical imagingand deep imaging depth from ultrasonic imaging, moreover, it is a noninvasiveimaging mode, so it attracts more and more attention and has become a new hotspot inthe field of biomedical imaging.The research of PAI focus on three aspects, the forward problem, the inverseproblem and the imaging system. The forward problem is the process that the tissueabsorbs the laser and emits the photoacoustic signal, the inverse problem also calledreconstruction process is the process that photoacoustic signal be received by the probeand be used to rebuild the absorption distribution by the computer, the research ofimaging system concentrates on the optimizing of the system parts such as the lasersource, the probe and so on.This paper mainly focuses on the optimization of the algorithm in the inverseproblem and the denoising of the photoacoustic signal. When studying the filterback-projection reconstruction algorithm existed, we found the reconstruction resultscontains artifact of some degree, then we introduce the anisotropic diffusion theoryinto reconstruction algorithm, and propose the algorithm filter back-projectionreconstruction algorithm based on the anisotropic diffusion theory, this new algorithmhas been proved to greatly reduce the effect of artifact. Furthermore, we consider thenoise that may be contained in the photoacoustic signal, we use some filters alreadyexisted to deal with the noise and find the result not good enough, so we propose themethod of combining some filters together also called the cascade filter, here wecombine three kinds of filter together and use them to filter out the noise, finally weget a much better result of denoising.