A Generative Model-Enhanced Deep Image Prior Framework for Unsupervised 3-D Electrical Impedance Tomography

View at publisher Summarise with VIRAT
Authors: Yuanbin Wu, Xinhao Bai, Jiewei Lu, Jianda Han, Ningbo Yu
Cited by: 0
References: 33

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Electrical impedance tomography (EIT), a non-invasive, radiation-free, and real-time imaging modality of tissue conductivity changes, has become an important clinical tool. Three-dimensional EIT offers superior spatial characterization of lung function, but its image reconstruction remains challenging due to the severely ill-posed inverse problem and high sensitivity to measurement noise. This paper presents a generative model-enhanced deep image prior framework for unsupervised three-dimensional electrical impedance tomography reconstruction, where the deep image prior provides implicit regularization and the generative model prior imposes explicit shape constraint to enforce physiologically plausible image. A dual-output network is designed to simultaneously reconstruct the conductivity image and update the generator latent variable. The reconstruction is performed through instance-wise unsupervised optimization with a unified objective that combines data-consistency loss and shape constraint, allowing both the network parameters and the latent representation to be jointly updated. In addition, a residual Noise2Noise denoiser module is devised for EIT data to suppress measurement noise in the voltage input by leveraging residual learning strategy and reciprocal voltage pairs. Extensive simulations and experiments are conducted to evaluate the reconstruction quality, generalization performance, and denoising ability under diverse pulmonary ventilation conditions. The results demonstrate the proposed framework significantly improves 3D pulmonary ventilation imaging with accurate conductivity distribution, high shape fidelity and effective artifact suppression.

Fields of Study & Subjects

Electrical and Bioimpedance Tomography Geophysical and Geoelectrical Methods Microwave Imaging and Scattering Analysis

Publication details

Journal / source
IEEE Transactions on Instrumentation and Measurement
Volume
75
Pages
1–15
Publication date
1 January 2026
Language
EN