Phase 3 · Clinical EMU & Electrophysiology Core Visit

Lab Visit: Clinical EMU & High-Density EEG Core

Clinical protocols for continuous long-term video-EEG in the Epilepsy Monitoring Unit (EMU), 128/256-channel high-density electroencephalography, Faraday cage isolation, electrode impedance physics, active shielding, and continuous bedside neonatal amplitude-integrated EEG (aEEG).

Course: IDNE 701 (Fall 2026) Date: Oct 23, 2026 (Week 9) Location: Clinical Epilepsy Monitoring Unit (EMU) & Electrophysiology Core Instructor: Mark Bolding, PhD

01. Clinical EMU Infrastructure & Video Telemetry

The Epilepsy Monitoring Unit (EMU) is a specialized inpatient facility designed for continuous, multi-day recording of electroencephalographic (EEG) and synchronized high-definition video telemetry. Its primary mission is capturing spontaneous paroxysmal events to localize epileptogenic seizure onset zones (SOZs) for pre-surgical evaluation and to differentiate epileptic seizures from non-epileptic psychogenic events.

📹 Time-Synchronized Video Telemetry

Dual pan-tilt-zoom (PTZ) infrared cameras continuous track patient motor behavior, eye deviation, and automations. Video frames are stamped with sample-accurate SMPTE hardware timecodes, perfectly synchronized to raw amplifier timebases.

🔘 Patient & Staff Event Triggers

Bedside pushbutton trigger inputs insert instantaneous digital event flags into the continuous data stream when the patient perceives an aura or nursing staff observe seizure semiology, initiating immediate high-rate video archiving.

🔌 Galvanic Isolation & Patient Safety

Headbox preamplifiers employ optical or magnetic digital isolators with isolation barrier ratings $> 4\,\text{kV}_{\text{RMS}}$. Chassis leakage current must remain $< 10\,\mu\text{A}$ (IEC 60601-1 Type BF/CF cardiac-floating safety rating).

🩹 Long-Term Skin Preservation

Unlike acute 30-minute diagnostic EEG, EMU monitoring lasts 72–168 hours. Collodion-affixed disc electrodes require hydrogel rehydration ports and pressure-relief silicone pads to prevent skin maceration and contact necrosis.

02. Electromagnetic Shielding & Driven Right Leg (DRL) Physics

Surface scalp potentials generated by cortical postsynaptic dipoles are microscopic ($10 - 100\,\mu\text{V}$). In contrast, ambient $120\,\text{V}/60\,\text{Hz}$ building powerlines couple capacitive displacement currents $I_d = C \frac{dV}{dt}$ into the human body, inducing common-mode voltages of $1 - 10\,\text{mV}$ ($40 - 60\,\text{dB}$ larger than the neural signal).

$$V_{\text{cm}} = \frac{1}{N}\sum_{i=1}^N V_i, \qquad V_{\text{DRL}} = -A_{\text{DRL}} \cdot V_{\text{cm}} \implies V_{\text{body, effective}} = \frac{V_{\text{cm}}}{1 + A_{\text{DRL}}}$$
Equation 1: Driven Right Leg (DRL) closed-loop common-mode cancellation

🛡️ Faraday Room Attenuation

Double-walled continuous copper-mesh suites attenuate ambient electromagnetic radiation by $> 80\,\text{dB}$ across $10\,\text{kHz} - 1\,\text{GHz}$. Penetration panels route isolated DC power and fiber-optic signals without compromising the Faradaic shield.

⚡ Active Coaxial Cable Shielding

Parasitic capacitance between electrode wires and ambient noise causes cable movement artifacts. Active unity-gain buffer amplifiers drive the outer coaxial cable shield at the exact same instantaneous potential as the inner signal wire, nullifying displacement current leakage.

03. Interactive Station 1: Electrode Impedance & Thermal Noise Budget

Measure and adjust contact impedance for 8 standard scalp montage locations. Observe how electrode-electrolyte contact resistance directly alters the Johnson-Nyquist thermal noise floor and $60\,\text{Hz}$ powerline common-mode conversion.

🎛️ Real-Time Electrode Impedance Checker Threshold: < 5.0 kΩ
Mean Impedance
3.2 kΩ
Max Impedance Mismatch
1.4 kΩ
Thermal Noise (0.5–70 Hz)
0.06 µV_rms
60 Hz Conversion
< 0.2 µV

04. Volume Conduction & The Forward Lead-Field Matrix

Scalp surface potentials $V_e$ represent the volume-conducted superposition of dipolar current sources $\vec{J}_p(\vec{r})$ across cortical pyramidal columns governed by Poisson's quasi-static volume conduction equation:

$$\nabla \cdot (\sigma(\vec{r}) \nabla \Phi(\vec{r})) = \nabla \cdot \vec{J}_p(\vec{r}) \implies \mathbf{v} = \mathbf{L}\mathbf{j} + \mathbf{n}$$
Equation 2: Poisson volume conduction and discrete lead-field matrix L mapping dipoles j to scalp voltages v
Anatomical Layer Adult Conductivity ($\sigma$) Neonatal Conductivity ($\sigma$) Spatial Filtering Effect
Scalp & Subcutis $0.33 - 0.43\,\text{S/m}$ $0.40 - 0.50\,\text{S/m}$ High water content; minimal distortion.
Cranial Bone (Skull) $0.0042 - 0.010\,\text{S/m}$ $0.025 - 0.045\,\text{S/m}$ Adult skull is heavily insulating ($\sim 1/80$ ratio), low-pass blurring dipoles. Neonatal skull is thin and 5–10× more conductive.
Fontanelles & Sutures N/A (Fused) $1.79\,\text{S/m}$ (CSF equivalent) Unfused anterior fontanelle provides a massive open electrical window, creating sharp localized potential peaks.
Cerebrospinal Fluid (CSF) $1.79\,\text{S/m}$ $1.79\,\text{S/m}$ Current shunting layer; spreads dipolar currents laterally before reaching bone.

05. Interactive Station 2: Bedside aEEG Seizure & Burst-Suppression Simulator

In the Neonatal Intensive Care Unit (NICU), full 128-channel montages are impractical due to extreme skin fragility and incubator nursing. Clinicians rely on amplitude-integrated EEG (aEEG) (typically biparietal C3–P3, C4–P4): signals are asymmetric bandpass filtered ($2 - 15\,\text{Hz}$), rectified, and compressed on a semi-logarithmic scale ($< 10\,\mu\text{V}$ linear, $> 10\,\mu\text{V}$ log) and written at slow speed ($6\,\text{cm/hr}$).

👶 NICU Bedside aEEG Monitor & Trace Classifier Continuous Normal Voltage (CNV)
COMPRESSED aEEG TREND (6 cm/hr, 2-Hour Window, Semi-Log Scale: 0–10 µV Linear, 10–100 µV Log)
RAW BIPARIETAL EEG STRIP (C3–P3, 10-Second Epoch, 250 Hz Sample Rate)
Upper Margin
28.4 µV
Lower Margin
11.2 µV
Burst-Suppression Ratio (BSR)
0 %
Clinical Alert Status
NORMAL

06. Grand Challenge 06 Focus: NICU Bedside Constraints

🛡️ Protecting Fragile Neonatal Skin & Temperature Calibration

Designing clinical neurotechnology for premature infants ($< 32$ weeks gestational age) imposes physical and chemical boundaries that differ drastically from adult ambulatory EMU systems:

  • Stratum Corneum Fragility: Preterm infants have only 2–3 layers of cornified skin cells. Standard abrasive skin-prep pastes (e.g. NuPrep) strip epidermis, leading to sepsis. Hydrogel electrolyte reservoirs with gentle saline activation must achieve $< 5\,\text{k}\Omega$ without mechanical abrasion.
  • Sustained Interface Pressure: Cap straps and head holders must distribute force evenly such that sustained pressure remains $< 20\,\text{mmHg}$ ($2.67\,\text{kPa}$), preserving capillary perfusion to prevent pressure ulcer necrosis.
  • 72-Hour Therapeutic Hypothermia ($33.5^\circ\text{C}$): Standard of care for neonatal hypoxic-ischemic encephalopathy (HIE) lowers body temperature to $306.65\,\text{K}$. This shifts electrode electrochemical half-cell potential, increases skin contact impedance, and suppresses background cerebral metabolic rate ($\text{CMRO}_2$).

07. Student Lab Observation Log & Verification Protocol

Complete the following checklist and qualitative observations during your hands-on tour of the Clinical EMU and Neuroengineering EEG Core. Save your progress locally or export your dossier report for your course design archive.