Bayesian-based weighted optoacoustic tomographic reconstruction in acoustic scattering media

The high optoacoustic resolution at depths beyond the diffusive limit of light stems from the low scattering of sound, as compared to photons, within biological tissues. However, some biological samples contain strongly mismatched tissues such as bones or lungs that generally produce acoustic reflections and scattering, and image distortion is consequently produced by assuming an acoustically homogeneous medium. We describe herein a statistical procedure to modify the reconstruction algorithms in order to avoid such distortion. The procedure is based on weighting the contribution of the collected optoacoustic signals to the reconstruction with the probability that they are not affected by reflections or scattering. A rough estimation of such probability by considering an area enclosing the sample allows significantly reducing the artefacts associated to acoustic distortion. Furthermore, the available structural information of the imaging sample can be incorporated in the estimation of the distortion probability, in a way that a further improvement in the quality of the reconstructed images is achieved. The benefit of the reconstruction procedure described herein is showcased by reconstructing tissue mimicking phantoms containing air-gaps. In all cases, the image artefacts produced when no weighting is done are significantly reduced.