Osteoarticular diseases are the most prevalent chronic pains and long-term disabilities with hundreds of millions people affected worldwide. They represent a major socioeconomic burden: a recent study estimated that the cost of arthritis, in industrialised countries, is ranging from 1% up to 2.5% of Gross National Product. Until now there is a lack of imaging modalities that can in a single image render and differentiate in 3D all the different tissues of a joint for pre-clinical or clinical studies. X-ray Phase Contrast Tomography emerged few decades ago in synchrotrons showing an unprecedented level of details for all the different biological tissue types in a single image.

Speckle tomography is a recently introduced Phase Contrast Imaging technique that, thanks to an extremely simple set-up, has been demonstrated to be easily transferable in 2D to conventional X-ray sources. Typically, a beam modulation is induced by a simple membrane containing absorbing or refracting grains. By image processing techniques it is then possible to retrieve absorption and refraction of the sample.

In this project, we propose to develop a new computational and experimental approach for the application of 3D Speckle tomography on human anatomical pieces using a conventional X-ray source. We will work on the reconstructed volume refraction and absorption variables. By working on the volume we will enforce the sparsity requirement on the solution and reduce the amount of information that needs to be processed from the measured projections and therefore reduce the radiation dose deposition.