Domain independent post-processing with graph U-nets: Applications to Electrical Impedance Tomographic Imaging

by   William Herzberg, et al.

Reconstruction of tomographic images from boundary measurements requires flexibility with respect to target domains. For instance, when the system equations are modeled by partial differential equations the reconstruction is usually done on finite element (FE) meshes, allowing for flexible geometries. Thus, any processing of the obtained reconstructions should be ideally done on the FE mesh as well. For this purpose, we extend the hugely successful U-Net architecture that is limited to rectangular pixel or voxel domains to an equivalent that works flexibly on FE meshes. To achieve this, the FE mesh is converted into a graph and we formulate a graph U-Net with a new cluster pooling and unpooling on the graph that mimics the classic neighborhood based max-pooling. We demonstrate effectiveness and flexibility of the graph U-Net for improving reconstructions from electrical impedance tomographic (EIT) measurements, a nonlinear and highly ill-posed inverse problem. The performance is evaluated for simulated data and from three measurement devices with different measurement geometries and instrumentations. We successfully show that such networks can be trained with a simple two-dimensional simulated training set and generalize to very different domains, including measurements from a three-dimensional device and subsequent 3D reconstructions.


page 1

page 5

page 7

page 8

page 9

page 10


Graph Convolutional Networks for Model-Based Learning in Nonlinear Inverse Problems

The majority of model-based learned image reconstruction methods in medi...

Geometrical discretisations for unfitted finite elements on explicit boundary representations

Unfitted (also known as embedded or immersed) finite element approximati...

Fractured Meshes

This work introduces “generalized meshes", a type of meshes suited for t...

phi-FEM for the heat equation: optimal convergence on unfitted meshes in space

Thanks to a finite element method, we solve numerically parabolic partia...

Learning the Solution Operator of Boundary Value Problems using Graph Neural Networks

As an alternative to classical numerical solvers for partial differentia...

MAgNET: A Graph U-Net Architecture for Mesh-Based Simulations

In many cutting-edge applications, high-fidelity computational models pr...

Please sign up or login with your details

Forgot password? Click here to reset