Optimization of the image reconstruction procedure in multi-focal photoacoustic computed tomography

Photoacoustic-computed microscopy (PACM) differs from conventional photoacoustic microscopy (PAM) imaging techniques in a way that thousands of optical foci are generated simultaneously using a two-dimensional microlens array, and raster-scanning these optical foci provides wide-field images. A major limitation of PACM is the slow imaging speed caused by the high power pulsed lasers and large amount of acoustic detectors. Here, we addressed this problem through compressed sensing and image inpainting. Compressed sensing minimizes the number of transducer elements used to acquire each frame, while inpainting minimizes the scanning steps. Combining these two approaches, we improved the imaging speed by sixteen times.