Optimal Transportation for Electrical Impedance Tomography

Gang Bao,Yixuan Zhang

DOI: https://doi.org/10.48550/arXiv.2210.16082

2022-10-28

Abstract:This work establishes a framework for solving inverse boundary problems with the geodesic based quadratic Wasserstein distance ($W_{2}$). A general form of the Fréchet gradient is systematically derived by optimal transportation (OT) theory. In addition, a fast algorithm based on the new formulation of OT on $\mathbb{S}^{1}$ is developed to solve the corresponding optimal transport problem. The computational complexity of the algorithm is reduced to $O(N)$ from $O(N^{3})$ of the traditional method. Combining with the adjoint-state method, this framework provides a new computational approach for solving the challenging electrical impedance tomography (EIT) problem. Numerical examples are presented to illustrate the effectiveness of our method.

Numerical Analysis,Mathematical Physics

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper aims to develop a method based on the quadratic Wasserstein distance ($W_2$) to solve severely ill - posed inverse problems, especially the Electrical Impedance Tomography (EIT) problem. The goal of the EIT problem is to determine the conductivity distribution of the medium from the voltage - current mapping on the boundary. Specifically:

1. **Challenges of the EIT problem**:
- The EIT problem is highly nonlinear and severely ill - posed, resulting in instability during the solution process.
- Traditional iterative optimization methods face problems of high computational complexity and sensitivity to noise when solving the EIT problem.

2. **Introduction of optimal transport theory**:
- The paper proposes a new framework, using the geodesic - based quadratic Wasserstein distance ($W_2$) as a metric to measure the differences in data matching.
- The $W_2$ distance has better convexity and robustness, and can capture both magnitude and spatial information simultaneously, having an advantage over the traditional $L_2$ metric.

3. **Improved computational method**:
- A fast algorithm is proposed, reducing the computational complexity from the traditional $O(N^3)$ to $O(N)$, significantly improving the computational efficiency.
- Combined with the adjoint - state method, a new numerical method is provided to solve the EIT problem.

4. **Application background**:
- The EIT problem has important value in practical applications such as medical imaging and fluid monitoring, especially in the two - dimensional case, which is mathematically challenging but very practical.

### Summary

By introducing the optimal transport theory and the quadratic Wasserstein distance, the paper provides a new numerical method to stably and efficiently solve the Electrical Impedance Tomography problem, solving the problems of high computational complexity and sensitivity to noise in traditional methods.