1: \begin{abstract}

2: We consider the hybrid problem of reconstructing the isotropic electric conductivity of a body $\Omega$ from interior Current Density Imaging data obtainable using MRI measurements. We only require knowledge of the magnitude $|J|$ of one current generated by a given voltage $f$ on the boundary  $\partial\Omega$. As previously shown, the corresponding voltage potential u in $\Omega$ is a minimizer of the weighted least gradient problem

3: 

4: \[u=\hbox{argmin} \{\int_{\Omega}a(x)|\nabla u|: u \in H^{1}(\Omega),

5: \ \ u|_{\partial \Omega}=f\},\]

6: with $a(x)= |J(x)|$. In this paper we present an alternating split Bregman algorithm for treating such least gradient problems, for $a\in L^2(\Omega)$ non-negative and  $f\in H^{1/2}(\partial \Omega)$. 

7: We give a detailed convergence proof by focusing to a large extent on the dual problem. This leads naturally to the alternating split Bregman algorithm. The dual problem also turns out to yield a novel method to recover the full vector field $J$ from knowledge of its magnitude, and of the voltage $f$ on the boundary. We then present several numerical experiments that illustrate the convergence behavior of the proposed algorithm.

8: 

9: %%Inside the inclusions the data vanishes identically, thus

10: %%determining their shape and location. In two dimensions, or in

11: %%higher dimensions if conductivities are know to be Lipschitz

12: %%continuous, these inclusions exhaust the regions in which the

13: %%interior data vanishes identically.

14: %In three dimensions or higher, for H\"{o}lder continuous

15: %conductivities, it is possible to have $|J|\equiv 0$ in regions

16: %which are neither superconductive nor insulating. We discuss some

17: %possibilities to identify those and the conductivity outside.

18: \end{abstract}

19: