Physics Informed Neural Networks for Electrical Impedance Tomography

Authors

Danny Smyl, Georgia Institute of Technology
Tyler N. Tallman, Purdue University
Laura Homa, University of Dayton
Chenoa Flournoy, University of Dayton
Sarah J. Hamilton, Marquette UniversityFollow
John Wertz, Air Force Research Laboratory

Document Type

Article

Publication Date

8-2025

Publisher

Elsevier

Source Publication

Neural Networks

Source ISSN

0893-6080

Original Item ID

DOI: 10.1016/j.neunet.2025.107410

Abstract

Electrical Impedance Tomography (EIT) is an imaging modality used to reconstruct the internal conductivity distribution of a domain via boundary voltage measurements. In this paper, we present a novel EIT approach for integrated sensing of composite structures utilizing Physics Informed Neural Networks (PINNs). Unlike traditional data-driven only models, PINNs incorporate underlying physical principles governing EIT directly into the learning process, enabling precise and rapid reconstructions. We demonstrate the effectiveness of PINNs with a variety of physical constraints for integrated sensing. The proposed approach has potential to enhance material characterization and condition monitoring, offering a robust alternative to classical EIT approaches.

Comments

Neural Networks, Vol. 188 (August 2025). DOI.

Recommended Citation

Smyl, Danny; Tallman, Tyler N.; Homa, Laura; Flournoy, Chenoa; Hamilton, Sarah J.; and Wertz, John, "Physics Informed Neural Networks for Electrical Impedance Tomography" (2025). Mathematical and Statistical Science Faculty Research and Publications. 149.
https://epublications.marquette.edu/math_fac/149