Model Based Iterative Reconstruction of Tomographic data with the Core Imaging Library #

Edoardo Pasca, Evelina Ametova, Gemma Fardell, Jakob Sauer Jørgensen, Laura Murgatroyd, Evangelos Papoutsellis

11:50 Tuesday in 2Q48.

Part of the Advances in applied numerical linear algebra and its applications session.

Abstract #

Computed Tomography (CT) is a powerful tool to study samples non- destructively. It is utilised in medical imaging, materials science, manufacturing, bio-science and security. The reconstruction of X-Ray CT images is typically done using the Filtered Back Projection (FBP) algorithm in cases where there is a simple acquisition geometry and we have full radial information. However, for complex systems, noisy or subsampled data the reconstruction can be done employing iterative methods to solve a model based optimisation problem. The Core Imaging Library (CIL) is a Python package for tomographic imaging reconstruction. CIL provides tools for loading, preprocessing and visualizing tomographic data as well as an extensive modular optimization framework for prototyping reconstruction methods including sparsity and non- smooth regularization.