Acoustic resolution photoacoustic microscopy (AR-PAM) provides high imaging resolution at an imaging depth beyond the optical diffusion limit. In these systems, the lateral resolution can be improved by using an ultrasound transducer (UST) with high numerical aperture (NA). However, increasing the NA of UST leads to a decreased depth of focus (DOF), resulting in deterioration of resolution in the out-of-focus regions. Image processing algorithms are commonly employed to improve the resolution outside the focal region. In this work, we propose a deep learning method to enhance the images obtained using AR-PAM. AR-PAM images were first simulated using k-wave toolbox in MATLAB. Convolutional neural network (CNN) based architecture was trained using these simulated images. The resulting model was then tested on experimentally collected AR-PAM images. Our results demonstrated that this method can significantly improve the outof-focus resolution of AR-PAM, thereby enhancing the image quality.
|