Development of an Anatomically Realistic Forward Solver for Thoracic Electrical Impedance Tomography

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Modeling

Background:

• Electrical impedance tomography (EIT) offers safe, low-cost patient monitoring during mechanical ventilation.
• Inconsistent reconstruction algorithms hinder cross-institutional interpretation of regional ventilation data.
• A unified thoracic EIT reconstruction algorithm requires a reliable forward model.

Purpose of the Study:

• To develop an anatomically realistic forward solver for thoracic EIT.
• To generate high-fidelity thoracic impedance data for algorithm development.
• To facilitate the creation of a consensual thoracic EIT reconstruction algorithm.

Main Methods:

• Constructed an anatomically realistic thoracic forward solver using high-resolution MRI data.
• Validated the solver by comparing predicted and observed surface impedance measurements.
• Assessed the accuracy of the forward solver's predictions.

Main Results:

• The developed forward solver demonstrated high fidelity in predicting thoracic impedance data.
• The relative error between predicted and observed impedance measurements was within 5%.
• The solver accurately models electric fields within the thoracic volume.

Conclusions:

• The presented forward solver is suitable for developing and evaluating thoracic EIT reconstruction algorithms.
• This tool can contribute to establishing a unified approach for thoracic EIT.
• Accurate forward modeling is crucial for advancing EIT in clinical ventilation monitoring.