Skull Modeling Effects in Conductivity Estimates Using Parametric Electrical Impedance Tomography

Area of Science:

• Biomedical Engineering
• Electrical Engineering
• Medical Imaging

Background:

• Accurate electrical conductivity values for human tissues are crucial for interpreting medical imaging data.
• Electrical impedance tomography (EIT) is a non-invasive imaging technique that measures electrical conductivity distributions.
• Previous EIT studies may have used simplified head models, potentially leading to inaccurate conductivity estimates, particularly for the skull.

Purpose of the Study:

• To estimate electrical conductivity of scalp, skull, compact bone, and marrow bone using electrical impedance tomography (EIT).
• To assess the impact of skull modeling simplifications on EIT-derived conductivity values.
• To provide updated conductivity estimates for widely used head models.

Main Methods:

• Collected EIT data using 62 current injection pairs from four subjects.
• Developed five finite element (FE) head models with varying degrees of skull simplification (6-8 million elements).
• Estimated electrical conductivity for scalp, skull, marrow bone, and compact bone tissues for each subject and model.

Main Results:

• Skull modeling simplifications (e.g., closed holes, simplified boundary elements, omitted CSF layer) led to conductivity overestimations of 10% to 70%.
• Average estimated conductivities: scalp (288 ± 53 mS/m), compact bone (4.3 ± 0.08 mS/m), and whole skull (5.5 ± 1.25 mS/m).
• Marrow bone conductivity estimates exhibited significant variability.

Conclusions:

• Current EIT-derived skull conductivity estimates are lower than previously reported literature values when using detailed skull models.
• Previous in vivo EIT studies likely overestimated skull conductivity due to the use of less complex head models.
• Recommended updated skull conductivity values (5.5 ± 1.25 mS/m) for detailed head models, replacing older literature values (7-10 mS/m).