⚡ electrical engineering

A Lumped-Element Electrical Model of the Human Head for Brain-Oriented Applications

This paper presents and validates a compact, frequency-dependent lumped-element RC circuit model of the human head's three-shell structure, demonstrating that accounting for tissue dispersion and capacitive pathways is essential to accurately predict scalp potentials and avoid overestimation.

Original authors: Angelo Faccia, Ermanno Citraro, Francesco P. Andriulli

Published 2026-05-29

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Angelo Faccia, Ermanno Citraro, Francesco P. Andriulli

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Building a "Circuit Board" for the Human Head

Imagine you are an engineer trying to design a device that either listens to brain signals (like a super-advanced hearing aid for thoughts) or sends gentle electrical pulses to the brain to help with movement or mood. To do this safely and effectively, you need to know exactly how electricity moves through the human head.

The problem is that the human head is complicated. It has three main layers: the soft brain, the hard skull, and the scalp. Electricity doesn't just flow straight through these layers; it spreads out, gets stuck, and changes behavior depending on how fast the electricity is moving (its frequency).

Usually, scientists use massive, super-computer simulations to figure this out. It's like trying to predict the weather by simulating every single water molecule in the atmosphere. It's accurate, but it takes too long and is too heavy to use when you are trying to quickly design a new electronic device.

The Solution: The authors of this paper created a "shortcut." They built a compact electrical circuit model (a simplified map) that acts like a stand-in for the whole head. Instead of simulating millions of points, they used a small network of resistors and capacitors (standard electronic parts) to mimic how the brain, skull, and scalp behave.

How the Model Works: The "Highway and Side Street" Analogy

To understand their model, imagine the head as a series of concentric onion layers (Brain, Skull, Scalp).

The Radial Path (The Highway): When electricity is generated deep in the brain, some of it tries to travel straight out toward the surface, like a car driving down a highway. The model uses a specific electrical component to represent this "through-layer" journey.
The Tangential Path (The Side Streets): However, electricity doesn't just go straight out. Some of it gets stuck or flows sideways within a layer before finding a way out, like a car taking a detour through side streets. The model adds a second set of components to represent this "sideways" flow.

By combining these "highways" and "side streets" for each layer, the model captures how electricity actually splits and moves, which simpler models often miss.

The Secret Sauce: Why "Frequency" Matters

The paper highlights a crucial detail: Electricity in the body isn't static.

Think of the tissues (brain, skull, scalp) not just as sponges that let water (current) through, but as sponges that change texture depending on how fast you pour the water.

Low frequency (slow pour): The tissue acts mostly like a resistor (a simple roadblock).
High frequency (fast pour): The tissue starts acting like a capacitor (a spring or a storage tank). It resists the flow differently and stores some energy.

The authors' model includes special "dispersive" components. This means the model knows that the "resistance" and "storage" of the skull and scalp change as the speed of the signal changes.

What They Found: The Danger of Ignoring the "Spring" Effect

The team tested their new circuit model against a very complex, math-heavy "gold standard" solution (which they call the "Semi-Analytical Reference").

The Result: Their simple circuit model matched the complex math almost perfectly across a wide range of speeds (from 10 Hz up to 50 kHz).

The Warning: They also tested what happens if you ignore the "spring" effect (the capacitive part) and the changing texture (dispersion).

The Mistake: If you treat the head as a simple, static resistor (like a plain piece of wire), your model will grossly overestimate the voltage on the scalp.
The Scale of Error: At higher frequencies, a simple model might predict a voltage that is more than double (over 100% error) what is actually happening. Even at lower frequencies, the error is significant.

Why This Matters (According to the Paper)

The authors state that this model is designed to be a surrogate—a lightweight stand-in for the heavy, complex simulations.

For Designers: It allows engineers to plug the head model directly into standard circuit simulators. This means they can rapidly prototype and test neuro-sensing or neuro-stimulation devices without waiting days for a supercomputer to finish a calculation.
For Safety: Because the model correctly accounts for the "spring" effect (capacitance) and changing resistance, it prevents engineers from designing devices that might be unsafe because they overestimated how much electricity reaches the skin.

Summary in One Sentence

The authors created a simple, fast-to-calculate electrical circuit that mimics the human head, proving that you must include "spring-like" electrical behaviors to avoid wildly wrong predictions about how much electricity reaches the scalp.

Technical Summary: A Lumped-Element Electrical Model of the Human Head for Brain-Oriented Applications

Problem Statement
Accurate estimation of electric potential and current density distributions in head tissues is critical for the design and safety assessment of neuro-sensing and neuro-stimulation systems (e.g., EEG source imaging, Deep Brain Stimulation). While numerical techniques like FEM, FDM, and BEM can model realistic anatomy, their high computational cost and limited compatibility with circuit simulators hinder rapid prototyping and real-time closed-loop evaluation. Existing circuit-oriented alternatives, such as resistor mesh models, often rely on purely resistive assumptions that neglect capacitive and dispersive effects. These omissions become increasingly significant as operating frequencies rise into the sub-MHz range, potentially leading to inaccurate predictions of scalp potentials.

Methodology
The authors propose a compact, lumped-element RC equivalent circuit (surrogate model) for the human head under electro-quasi-static (EQS) assumptions. The methodology involves the following steps:

Geometry and Physics: The model utilizes a canonical three-shell spherical geometry (brain, skull, scalp) embedded in an infinite air region. It accounts for frequency-dependent tissue conductivity ( $\sigma(\omega)$ ) and permittivity ( $\varepsilon(\omega)$ ), treating tissue behavior through complex, dispersive conductivity.
Circuit Topology: Unlike simple series stacking, the model employs a specific topology for each tissue layer consisting of:
- Radial Impedance ( $Z^r_i$ ): Models current flow through the layer.
- Tangential Impedance ( $Z^\theta_i$ ): Models intra-layer return currents.
- Implementation: Each branch is realized as a parallel RC element. Resistances scale inversely with conductivity, and capacitances scale with permittivity.
- Boundary Conditions: Kirchhoff's Current Law (KCL) at radial nodes enforces the continuity of total current density, naturally associating interfacial surface charge accumulation with capacitive behavior.
Parameterization:
- Geometry: A dimensionless coefficient ( $\gamma_i$ ) couples tangential and radial pathways, parameterized as polynomial functions of radius ratios to capture morphological variability.
- Source Modeling: An intracranial current dipole is modeled as a current source. To handle off-center (eccentric) dipoles, the brain's radial impedance is split into two asymmetric branches ( $Z^{r_{up}}_{brain}$ and $Z^{r_{down}}_{brain}$ ) controlled by an eccentricity-dependent parameter $\alpha(\eta)$ .
Validation Strategy: The model is validated against a semi-analytical reference solution based on scalar spherical harmonics (SSH). The validation covers multiple geometric configurations (varying skull thickness and dipole eccentricity) and a frequency range from 10 Hz to 50 kHz.

Key Contributions

Compact Surrogate Circuit: Development of a minimal RC network that reproduces the peak scalp potential generated by an intracranial dipolar source up to 50 kHz.
Dispersive Modeling: Explicit integration of frequency-dependent tissue properties and displacement-current pathways via capacitive branches, addressing the limitations of purely resistive models.
Eccentricity Handling: A method to model cortical source locations with varying eccentricity by asymmetrically redistributing brain radial impedances.
Reciprocity Compatibility: The model is designed to support both sensing and stimulation applications via the reciprocity theorem.

Results

Agreement with Reference: The lumped model demonstrates good agreement with the SSH reference solution. The Mean Relative Frequency Error (MRFE) for peak scalp potential increases moderately with dipole eccentricity and skull thickness variations but remains low overall. The highest sensitivity is observed for highly eccentric dipoles.
Impact of Dispersion and Capacitance: Comparative analysis of three circuit configurations (purely ohmic, dispersive conductivity only, and full dispersive RC) reveals significant errors when capacitive and dispersive effects are ignored:
- Purely Ohmic: Leads to substantial overestimation of scalp voltage, with errors exceeding 100% at the upper end of the frequency band (50 kHz) and becoming appreciable even at tens of hertz.
- Dispersive Conductivity Only: Reduces error significantly but still leaves discrepancies above 10% (peaking at 17% near 1.7 kHz).
- Full Dispersive RC: Required to minimize error, confirming that accurate modeling below 50 kHz necessitates accounting for both frequency-dependent impedance and capacitive current pathways.

Significance and Claims
The paper claims that the proposed circuit provides a practical surrogate model that balances computational efficiency with physical accuracy. By preserving the relevant physics of the EQS regime within a compact circuit netlist, it enables rapid device prototyping and circuit-level integration for neuro-engineering applications. The authors emphasize that neglecting dispersion and capacitive pathways leads to systematic overestimation of scalp potentials, highlighting the necessity of their dispersive RC approach for broadband behavioral analysis. The model is presented as a foundation for future work involving tangential dipoles, anisotropic conductivities, and more realistic head geometries while maintaining circuit compactness.