Shape-Driven Difference Electrical Impedance Tomography.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Electromagnetics

Background:

• Electrical Impedance Tomography (EIT) is a non-invasive imaging modality.
• Conventional EIT reconstruction often relies on linear approximations.
• Accurate reconstruction is crucial for biomedical applications like lung imaging.

Purpose of the Study:

• To develop and evaluate a novel shape-driven reconstruction approach for difference EIT.
• To improve image quality and robustness in EIT, particularly for lung imaging.
• To incorporate geometric prior information directly into the EIT reconstruction process.

Main Methods:

• Formulated EIT reconstruction as a shape reconstruction problem using an explicit, geometrical methodology.
• Represented inclusion geometry with a shape and topology description function (STDF).
• Utilized moving morphable components (MMCs) as basic building blocks for embedded inclusions.

Main Results:

• The shape-driven approach demonstrated significantly improved image quality compared to the conventional linear method.
• The method showed tolerance to modeling errors, including background conductivity inhomogeneity.
• The approach proved robust against uncertainties in contact impedances and reference states.

Conclusions:

• The proposed shape-driven EIT reconstruction offers superior performance for lung imaging.
• This novel methodology enhances flexibility and accuracy by integrating geometric information.
• The approach provides a robust and improved imaging solution for EIT applications.