Electrical Impedance Tomography for Anisotropic Media: a Machine Learning Approach to Classify Inclusions

2025-02-06

AI Summary

This study addresses the inverse problem of electrical impedance tomography (EIT) in anisotropic media, aiming to detect and characterize single or multiple heterogeneous inclusions—assessing their existence, quantity, size, and internal conductivity anisotropy—solely from boundary electrostatic measurements (i.e., the Dirichlet-to-Neumann map) acquired via a limited number of electrodes (e.g., 16). We propose a novel hybrid modeling paradigm integrating artificial neural networks (ANNs) and support vector machines (SVMs): ANNs perform regression for inclusion size estimation, while SVMs handle multi-inclusion detection and anisotropy classification. Only two independent boundary excitations are required for high-accuracy prediction. The method eliminates reliance on dense boundary sampling or strong prior models. Validation on both synthetic and experimental data demonstrates superior performance: high inclusion detection accuracy, significantly improved anisotropy classification accuracy, and low size estimation error.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Intelligent Robots: State Estimation

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

We consider the problem in Electrical Impedance Tomography (EIT) of identifying one or multiple inclusions in a background-conducting body $Omegasubsetmathbb{R}^2$, from the knowledge of a finite number of electrostatic measurements taken on its boundary $partialOmega$ and modelled by the Dirichlet-to-Neumann (D-N) matrix. Once the presence of one inclusion in $Omega$ is established, our model, combined with the machine learning techniques of Artificial Neural Networks (ANN) and Support Vector Machines (SVM), may be used to determine the size of the inclusion, the presence of multiple inclusions, and also that of anisotropy within the inclusion(s). Utilising both real and simulated datasets within a 16-electrode setup, we achieve a high rate of inclusion detection and show that two measurements are sufficient to achieve a good level of accuracy when predicting the size of an inclusion. This underscores the substantial potential of integrating machine learning approaches with the more classical analysis of EIT and the inverse inclusion problem to extract critical insights, such as the presence of anisotropy.

Problem

• Identify inclusions in anisotropic media
• Classify inclusions using machine learning
• Determine size and anisotropy of inclusions

Innovation

• Machine learning enhances EIT accuracy
• ANN and SVM classify inclusions effectively
• Two measurements predict inclusion size reliably