Electrical Impedance Tomography Techniques in Biomedical Imaging

Electrical impedance tomography (EIT) is a non-invasive, non-ionising imaging modality that reconstructs the internal conductivity distribution of biological tissues by injecting imperceptible currents through surface electrodes and measuring the resulting boundary voltages. The core challenge of EIT lies in solving a highly ill-posed inverse problem: small measurement errors can lead to large uncertainties in the reconstructed images. Over the past decades, advances have encompassed robust finite-element forward models, multifrequency data acquisition to exploit tissue dispersion properties, and mathematically rigorous direct methods such as D-bar algorithms that apply nonlinear Fourier transforms. More recently, hybrid approaches have emerged, combining classical reconstruction schemes with machine-learning-based regularisation or unsupervised neural priors to enhance spatial resolution and suppress artefacts. These developments have broadened the clinical applicability of EIT, enabling real-time monitoring of lung ventilation, early detection of stroke subtypes, and dynamic imaging of organ perfusion. The portability, low cost and bedside compatibility of EIT systems render them particularly valuable for critical-care settings and point-of-care diagnostics in resource-limited environments.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Electrical Impedance Tomography Techniques in Biomedical Imaging publication trend

The graph below shows the total number of articles in electrical impedance tomography techniques in biomedical imaging across all publications each year (not limited to Nature Index journals).

Technical terms

Electrical impedance tomography (EIT): An imaging technique that infers internal conductivity distributions from surface current injections and voltage measurements.

Inverse problem: A mathematical problem in which internal properties are inferred from indirect external measurements, often ill-posed and sensitive to noise.

Deep image prior: A reconstruction strategy that uses the structure of a randomly initialised neural network as a regulariser without prior training data.

D-bar method: A direct reconstruction algorithm based on nonlinear Fourier analysis that transforms boundary data into conductivity images via a complex-valued scattering transform.

References

1. DeepEIT: Deep Image Prior Enabled Electrical Impedance Tomography. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023).
2. Solving Inverse PDE Problems using Grid-Free Monte Carlo Estimators. ACM Transactions on Graphics (2024).
3. Beltrami-net: domain-independent deep D-bar learning for absolute imaging with electrical impedance tomography (a-EIT). Physiological Measurement (2019).
4. Stroke type differentiation using spectrally constrained multifrequency EIT: evaluation of feasibility in a realistic head model. Physiological Measurement (2014).
5. Multifrequency Electrical Impedance Tomography Using Spectral Constraints. IEEE Transactions on Medical Imaging (2013).
6. A Fast Parallel Solver for the Forward Problem in Electrical Impedance Tomography. IEEE Transactions on Biomedical Engineering (2014).