1. Overview: Neurodynamics of Epileptiform Synchronization
Epilepsy represents one of the most widespread chronic neurological disorders, characterized by recurrent, unprovoked paroxysmal hypersynchronous discharges across distributed neuronal ensembles. In the healthy brain, cortical equilibrium is maintained through a delicate homeostatic balance between excitatory post-synaptic potentials (EPSPs), mediated primarily by glutamatergic $\alpha$-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ($\text{AMPA}$) and $N$-methyl-D-aspartate ($\text{NMDA}$) receptors, and inhibitory post-synaptic potentials (IPSPs), controlled by $\gamma$-aminobutyric acid ($\text{GABA}_A$ and $\text{GABA}_B$) interneuronal populations.
During an ictogenic transition, this inhibitory-excitatory equilibrium suffers catastrophic failure. Thousands of pyramidal neurons within an epileptogenic zone enter a state of continuous paroxysmal depolarization shift (PDS), where high-frequency bursts of action potentials breach normal functional anatomical boundaries and recruit adjacent or distant connected networks.
Mathematically, seizure onset and termination can be conceptualized as bifurcations within multi-scale non-linear dynamical systems. Using the canonical Epileptor model developed by Jirsa et al., the transition between resting state, seizure oscillations, and postictal suppression is expressed as a set of five coupled non-linear differential equations operating across three distinct temporal scales:
$$\begin{aligned}
\dot{x}_1 &= y_1 - f_1(x_1, x_2) - z + I_{\text{ext1}} \\
\dot{y}_1 &= y_0 - 5x_1^2 - y_1 \\
\dot{z} &= \frac{1}{\tau_0} \Big( 4(x_1 - x_0) - z \Big) \\
\dot{x}_2 &= -y_2 + x_2 - x_2^3 + I_{\text{ext2}} + 0.002 g(x_1) - 0.3(z - 3.5) \\
\dot{y}_2 &= \frac{1}{\tau_2} \Big( -y_2 + f_2(x_2) \Big)
\end{aligned}$$
where $x_1$ represents the rapid ictal wavefront, $x_2$ governs slower spike-and-wave discharges, and $z$ acts as a permissive slow metabolic variable (representing intracellular sodium accumulation, extracellular potassium accumulation, or glial exhaustion) driving the ensemble through saddle-node and homoclinic bifurcations.
2. Operational Guidelines & Telemetry Workflow
This interactive laboratory provides a multi-modal sandbox for exploring spatial wavefront propagation, 3D anatomical localization, and dual-channel electroencephalographic (EEG) signatures across distinct clinical seizure classifications:
- Initializing a Seizure Sequence: Select a classification from the Classification Architecture dropdown and click "⚡ START SEQUENCE". The simulation initiates a calibrated clinical sequence:
- Preictal Stage (0.0s – 2.0s): Baseline instability, localized micro-spikes, and regional hyper-synchrony.
- Active Discharge (2.0s – 9.0s): High-voltage ictal paroxysms, rapid wavefront recruitment across connectome tracts, and maximal multi-channel EEG amplitude.
- Postictal Suppression (9.0s – 13.0s): Complete electrical flattening and neural fatigue characterized by slow restorative baseline recovery.
- Interactive Focal Point Selection: When Focal Seizure is active, click anywhere directly on the cortical mesh. The engine performs an optical raycast intersection, repositioning the illuminated focal beacon and redirecting the spatial propagation vector in real time.
- Anatomical Lobe Semiology: Hover and click on the floating 3D perimeter labels (Frontal, Parietal, Temporal, Occipital) to inspect clinical notes on regional functions, aura manifestations, and semiological motor presentations.
- Auditory Sonification Profiles: Enable "🔊 SOUND ON" and choose from three real-time audio profiles:
- Sub-Bass Frequency Hum: Low-pass filtered resonant drone tracking global cortical kinetic energy.
- Clinical Spike Clicks: High-frequency impulses mapping individual high-voltage spike discharges.
- Synchronous Rhythmic Sweeps: Resonant frequency sweeps illustrating widespread phase synchronization.
- Automated Audio-Driven Guided Tour: Click "▶ START DEMO" to start an automated, narrated sequence that tours each classification with coordinated HUD highlights and parameter transitions. Clicking any manual control immediately restores full manual control.
3. Mathematical Derivations & Rendering Engine Architecture
The rendering architecture integrates a WebGL pipeline via Three.js with an HTML5 2D Canvas oscilloscope. The human brain model is procedurally generated using 6,000 cortical nodes, subcortical thalamic nuclei, cerebellar folia, and descending brainstem tracts:
$$I_{\text{syn}}(t) = g_{\text{syn}} \sum_{j} (v(t) - E_{\text{syn}}) \cdot e^{-(t - t_j)/\tau_{\text{decay}}}$$
Wavefront propagation between cortical nodes is governed by distance-dependent spatial attenuation combined with anisotropic conduction velocities along simulated white-matter streamlines:
$$\Psi(\mathbf{r}, t) = A_0 \cdot \exp\left(-\frac{\|\mathbf{r} - \mathbf{r}_0\|^2}{2\sigma_{\text{spatial}}^2}\right) \cdot \cos\left(\omega t - \mathbf{k} \cdot (\mathbf{r} - \mathbf{r}_0)\right)$$
where $\mathbf{r}_0$ is the focal position vector, $\sigma_{\text{spatial}}$ dictates spatial recruitment breadth, and $\mathbf{k}$ defines the wavevector aligned with the connectome tracts.
Multi-Channel Electroencephalogram Forward Problem
The live EEG panel calculates simulated electrical dipole potentials at scalp electrode positions (left hemisphere Channel 1, right hemisphere Channel 2) using a simplified 4-sphere head volume conductor model:
$$V_e(t) = \sum_{i=1}^{N} \frac{1}{4\pi\sigma_c} \frac{\mathbf{p}_i(t) \cdot (\mathbf{r}_e - \mathbf{r}_i)}{\|\mathbf{r}_e - \mathbf{r}_i\|^3} + \xi(t)$$
where $\mathbf{p}_i(t)$ represents the current dipole moment of the $i$-th cortical pyramidal column, $\mathbf{r}_e$ is the electrode coordinate, $\sigma_c$ is skull conductivity, and $\xi(t)$ is pink ($1/f$) background physiological noise.
4. Future Directions: Clinical fMRI Integration & Closed-Loop Neuromodulation
The integration of real-time biophysical simulations into computational neurology offers practical opportunities for pre-surgical evaluation and neuro-technological interventions:
- Patient-Specific Tractography (DTI): Loading patient Diffusion Tensor Imaging (DTI) streamline tractographies to constrain electrical wavefronts along validated anatomical white-matter axon bundles.
- Closed-Loop Responsive Neurostimulation (RNS): Simulating implantable intracranial electrocorticography (ECoG) electrodes that trigger high-frequency counter-stimulation pulses when preictal high-frequency oscillations (HFOs, 80–500 Hz) are detected.
- Multi-Channel Clinical Montages: Expanding the dual-trace oscilloscope into standard 16-channel bipolar double-banana configurations with real-time Fast Fourier Transform (FFT) power spectrum overlays.
Explore Related Neurotech Environments
Open Access License: This interactive educational module is released under
CC BY-NC 4.0 (Attribution-NonCommercial)
for non-commercial research, academic study, and clinical education.
Commercial & Enterprise Licensing: For white-labeling, proprietary LMS/course embedding, hardware dashboard telemetry integration, or custom feature engineering, secure a commercial license at
BioniCloud.com or contact
Dr. Yuri Beno.