eprintid: 10139975 rev_number: 15 eprint_status: archive userid: 608 dir: disk0/10/13/99/75 datestamp: 2021-12-09 15:24:33 lastmod: 2021-12-09 15:25:13 status_changed: 2021-12-09 15:25:13 type: article metadata_visibility: show creators_name: Herzberg, W creators_name: Rowe, D creators_name: Hauptmann, A creators_name: Hamilton, S title: Graph Convolutional Networks for Model-Based Learning in Nonlinear Inverse Problems ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 keywords: Finite element method, graph convolutional networks, model-based deep learning, conductivity, electrical impedance tomography. note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: The majority of model-based learned image reconstruction methods in medical imaging have been limited to uniform domains, such as pixelated images. If the underlying model is solved on nonuniform meshes, arising from a finite element method typical for nonlinear inverse problems, interpolation and embeddings are needed. To overcome this, we present a flexible framework to extend model-based learning directly to nonuniform meshes, by interpreting the mesh as a graph and formulating our network architectures using graph convolutional neural networks. This gives rise to the proposed iterative Graph Convolutional Newton-type Method (GCNM), which includes the forward model in the solution of the inverse problem, while all updates are directly computed by the network on the problem specific mesh. We present results for Electrical Impedance Tomography, a severely ill-posed nonlinear inverse problem that is frequently solved via optimization-based methods, where the forward problem is solved by finite element methods. Results for absolute EIT imaging are compared to standard iterative methods as well as a graph residual network. We show that the GCNM has strong generalizability to different domain shapes and meshes, out of distribution data as well as experimental data, from purely simulated training data and without transfer training. date: 2021-12-02 date_type: published publisher: Institute of Electrical and Electronics Engineers (IEEE) official_url: http://dx.doi.org/10.1109/TCI.2021.3132190 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1908768 doi: 10.1109/tci.2021.3132190 lyricists_name: Hauptmann, Andreas lyricists_id: AHAUP49 actors_name: Hauptmann, Andreas actors_id: AHAUP49 actors_role: owner full_text_status: public publication: IEEE Transactions on Computational Imaging citation: Herzberg, W; Rowe, D; Hauptmann, A; Hamilton, S; (2021) Graph Convolutional Networks for Model-Based Learning in Nonlinear Inverse Problems. IEEE Transactions on Computational Imaging 10.1109/tci.2021.3132190 <https://doi.org/10.1109/tci.2021.3132190>. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10139975/1/2021_Herzberg_preprint.pdf