Compressive sampling in photo-acoustic imaging

Photo-acoustic (PA) imaging has been developped for different purposes but recent years has seen the technique gain interest with applications to small animal imaging. As a technique it is sensitive to endogenous optical contrast present in tissues and, contrary to diffuse optical imaging, it promises to bring high resolution imaging for in vivo studies at mid-range depths (3mm-10mm). However, a typical acquisition for the reconstruction of one slice in PA tomography can take up to approximatively 30 minutes which is clearly prohibitive for 3D imaging. This paper suggests a new reconstruction strategy using the compressive sampling formalism which states that a small number of linear projection of a compressible image contains enough information for reconstruction. This new scheme allows perfect reconstruction of numerical phantoms with only a fraction of the measurements normally needed with classical methods such as the pseudo-inverse.