A Parametric Level Set-Based Approach to Difference Imaging in Electrical Impedance Tomography.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Electromagnetics

Background:

• Electrical Impedance Tomography (EIT) is a non-invasive imaging technique.
• Conventional difference imaging in EIT reconstructs conductivity changes using surface measurements before and after an event.
• Existing methods face challenges with accuracy and computational load.

Purpose of the Study:

• To develop a novel difference imaging approach for EIT using parametric level sets (PLS).
• To improve the reconstruction of conductivity changes by accurately representing anomaly geometries.
• To enhance image quality and computational efficiency compared to conventional methods.

Main Methods:

• Utilized a parametric level set (PLS) method with Gaussian radial basis functions (GRBFs) for shape representation.
• Assumed piecewise constant conductivity changes for reconstruction.
• Validated the approach using simulations, phantom studies, and in vivo pig data.

Main Results:

• The PLS-based approach demonstrated flexibility in describing various anomaly shapes with fewer unknowns.
• Achieved significant improvements in image quality compared to conventional linear methods.
• Showed increased tolerance to modeling errors.

Conclusions:

• The proposed PLS-based difference imaging method offers a robust and efficient alternative for EIT.
• This technique enhances the accuracy and quality of conductivity change imaging.
• The method holds promise for various biomedical applications requiring precise EIT analysis.