Large volumetric optical-resolution photoacoustic microscopy with image fusion based on CNN feature extraction

Abstract

Photoacoustic imaging(PAI) is a emerging powerful and efficient imaging technology. Optical-resolution photoacoustic microscopy is an useful photoacoustic imaging technique combining the advantages of both optical imaging and acoustic imaging which obtains many attractive advanges such as high resolution, high contrast and so on. Laser beam is often focused strongly to achieve high resolution. However, this will lead to a poor depth-of-field and less structural information which limits the further application of this technology. Aimage fusion method based on CNN feature extraction is proposed to achieve large volumetric optical-resolution photoacoustic microscopy. First, two groups of simulated 3D photoacoustic data of different focal locations were obtained through photoacoustic microscopic imaging platform. Then B-scan data were fused and maximum projection of the reconstructed 3D data is taken to display the photoacoustic information. By comparing the source images and the fused image, we show that the proposed method can be implemented to obtain large volumetric and high-resolution photoacoustic images.