Blind-deconvolution Airy-beam photoacoustic microscopy for removing the sidelobe

Abstract

As an emerging nondestructive imaging technology, photoacoustic imaging (PAI), which is based on photoacoustic effect, combines the advantages of the high resolution and contrast of optical imaging and the high penetration depth of acoustic imaging. Thereinto, as a branch of photoacoustic imaging, photoacoustic microscopy inherited the advantages of photoacoustic imaging. The unique focusing mode of photoacoustic microscopy can meet the requirements of higher resolution in biological imaging and it has gained extensive applications in medical science field. In our previous work, in order to solve the shortcoming of traditional photoacoustic microscopy with a small depth of field, we have built a simulation platform for Airy-beam photoacoustic microscopy based on K-Wave MATLAB toolbox, which uses Airybeam to excite the initial photoacoustic signal. As non-diffraction beam, Airy-beam features the capacity of large depth of field. However, Airy-beam has sidelobe, which makes the edges of the image blurred. Convolution is a mathematical method of integrating transformations. The imaging result of optical system is the result of convolution of the sample and PSF of the system. PSF convolves with the sample, resulting in blurred image and reduced image resolution. The image can be restored by appropriate deconvolution techniques. In this paper, in order to weaken the influence of Airybeam’s sidelobe on the imaging results, Lucy-Richardson (LR) Algorithm is used to deconvolve the imaging results to obtain clearer restored images. LR Algorithm is a nonlinear iterative restoration algorithm based on Bayesian conditional probability model, and it is assumed that pixels in fuzzy images meet Possion distribution, and its optimal estimation criterion is maximum likelihood criterion. Using LR Algorithm for Airy-beam photoacoustic microscopy can greatly improve the system resolution and clearer imaging results can be obtained.