Monitoring Autonomic Tone During Spinal Cord Neuromodulation Using Wearble AURIS Sensor

This study validates a novel, non-invasive in-ear AURIS sensor utilizing a PDMS substrate for monitoring heart rate variability during spinal cord neuromodulation, demonstrating gold-standard fidelity comparable to conventional chest electrodes and establishing a robust foundation for closed-loop bioelectronic medicine.

The Gist

Imagine your body has a silent, automatic pilot system called the autonomic nervous system. It controls things you don't think about, like your heart rate, breathing, and blood pressure. Doctors are developing new "remote controls" (neuromodulation) to fix this system when it gets stuck or broken, but there's a big problem: we don't have a good way to watch the dashboard while we're driving.

Currently, to see if the remote control is working, doctors have to stick messy, sticky electrodes all over a patient's chest. It's like trying to tune a radio while someone is constantly slapping the antenna. The wires get loose, the sticky gel dries out, and if the patient moves even a little, the signal gets fuzzy.

This paper introduces a clever new solution called AURIS. Think of it as a high-tech earbud that doesn't just play music, but listens to your heart.

The Problem with the Old Way

Traditionally, to monitor the heart during these treatments, doctors use Ag/AgCl chest electrodes.

• The Analogy: Imagine trying to take a clear photo of a hummingbird while it's flying, but you have to hold the camera with a wobbly, sticky hand that gets sweaty and slips.
• The Issue: These chest stickers are uncomfortable, they dry out, and they pick up a lot of "noise" (static) when the patient moves or when the treatment (ultrasound) is happening.

The AURIS Solution: The "Earbud" Dashboard

The researchers created a new sensor that fits snugly inside the ear canal.

• The Material: It's made of a special, squishy, biocompatible material (PDMS) mixed with a conductive polymer. Think of it like a custom-molded earplug made of soft, conductive rubber that hugs the ear perfectly.
• Why the Ear? The ear canal is a quiet, stable tunnel. It's shielded from the body's movements (like the chest heaving when you breathe or the ultrasound vibrating the ribs).
• The Magic: Because it's so stable, this "earbud" can hear the heart's electrical signals almost as clearly as the messy chest stickers, but without the wobble or the mess.

The Experiment: Testing the Earbud

The team tested this on three rats. They did two things at the same time:

1. They stuck the traditional, messy electrodes on the rats' chests (the "Gold Standard").
2. They placed the new AURIS sensors in the rats' ears.

Then, they used Focused Ultrasound (FUS) to zap the spinal cord. This is like sending a precise sound wave to "reset" the autonomic nervous system. They wanted to see if the ear sensors could catch the heart's reaction to this zap just as well as the chest sensors.

The Results: A Perfect Match

The results were impressive.

• The Heartbeat: The ear sensors and the chest sensors agreed almost perfectly. The difference in heart rate was tiny (about 6 beats per minute), and the timing of the heartbeats was nearly identical.
• The "Complexity" Score: Heart rate isn't just about speed; it's about variability (how much the beat changes from second to second). This variability tells us how healthy the nervous system is. The AURIS sensors were excellent at detecting these subtle, complex changes, even when the chest sensors were struggling with noise.
• The Verdict: Statistically, the ear sensors were just as good as the chest sensors. They didn't get confused by the ultrasound or the rats moving around.

Why This Matters

This is a game-changer for the future of medicine.

• No More Sticky Pads: Patients won't have to deal with drying gels or itchy adhesives.
• Better Data: Because the ear sensors are so stable, doctors can get a clearer picture of how well a treatment is working in real-time.
• The Future: Imagine a future where you wear a comfortable earbud during a therapy session. The doctor can watch your "autonomic dashboard" in real-time and adjust the treatment instantly, like a pilot tweaking the autopilot based on live weather data.

In short: The researchers built a "smart earbud" that listens to your heart better than the old chest stickers, making it easier, cleaner, and more accurate to test new treatments for nervous system disorders. It's a small step for an earbud, but a giant leap for non-invasive medicine.

Technical Summary

1. Problem Statement

The transition of bioelectronic medicine from research to clinical application is currently hindered by a lack of non-invasive, high-fidelity sensors capable of monitoring autonomic tone during active neuromodulation.

• Limitations of Current Standards: Traditional monitoring relies on Mean Arterial Pressure (MAP) and chest-based Ag/AgCl ECG electrodes.
  • MAP: Often a lagging indicator of neural state and requires invasive arterial catheterization for high temporal resolution.
  • Chest ECG: Requires electrolytic gels (which evaporate), adhesives (causing skin irritation), and is highly susceptible to motion artifacts during active stimulation.
• The Gap: There is a critical need for a wearable sensor platform that offers "gold-standard" signal fidelity comparable to chest electrodes but with the durability, comfort, and motion resistance required for real-time clinical trials and closed-loop feedback systems.

2. Methodology

The study introduces and validates the AURIS platform, a novel in-ear sensing system.

• Sensor Design & Fabrication:
  • Material: Utilizes a flexible Polydimethylsiloxane (PDMS) substrate blended with PEDOT:PSS (a conductive polymer) to ensure biocompatibility and superior conformability.
  • Form Factor: Custom electrodes fabricated using pipette-tip casting, filled with conductive gel to minimize resistance.
  • Placement: Bilateral insertion into the auditory canals, leveraging the stable, low-impedance environment of the ear to minimize motion-induced noise and block auditory confounds (e.g., startle responses to ultrasound).
• Experimental Model:
  • Subjects: Adult female Sprague-Dawley rats ($n=3$).
  • Intervention: Focused Ultrasound (FUS) stimulation applied to the thoracic spinal cord (T12–T13) to modulate autonomic function. Both invasive (laminectomy) and non-invasive (skin incision only) delivery methods were tested.
  • Parameters: 500 kHz frequency, 250 mVpp amplitude, with stepwise increases in stimulation.
• Data Acquisition & Processing:
  • Comparison: AURIS (in-ear) signals were compared against gold-standard Ag/AgCl chest electrodes (Enthoven lead I).
  • Hardware: Tucker-Davis Technologies (TDT) RZ5D system for 1 kHz sampling.
  • Signal Processing: A custom Python pipeline extracted R-R intervals independently from both modalities. Metrics included time-domain (Mean RR, RMSSD, SDNN), frequency-domain (LF, HF, Total Power), and non-linear complexity measures (Sample Entropy, DFA $\alpha$ ratios, SD1/SD2).
  • Statistical Analysis: Independent t-tests for modality agreement; paired t-tests and Cohen's $d$ for sequential analysis (Stimulation vs. Washout periods).

3. Key Contributions

• Novel Sensor Architecture: Development of the AURIS in-ear sensor using a PDMS/PEDOT:PSS blend, offering a durable, non-invasive alternative to chest leads.
• Validation of Fidelity: Demonstration that in-ear sensors can achieve signal quality comparable to invasive chest electrodes during active neuromodulation.
• Complexity Metric Sensitivity: Identification that non-linear complexity metrics (specifically SD1/SD2 and DFA $\alpha$ ratios) are more robust indicators of autonomic shifts in small sample sizes than traditional time-domain metrics.
• Framework for Closed-Loop Systems: Establishing a technical foundation for using real-time HRV from wearable sensors to dynamically adjust neuromodulation parameters, moving away from invasive hemodynamic monitoring.

4. Key Results

• Sensor Agreement (Modality Validation):
  • The AURIS sensor showed no statistically significant difference compared to chest electrodes across all metrics (all $p > 0.46$).
  • Mean Heart Rate Difference: 6.03 BPM.
  • Mean RR Interval Delta: 3.18 ms.
  • Visual analysis (Poincaré plots and violin plots) confirmed that HR distributions and geometric patterns were virtually indistinguishable between ear and chest sites.
• Sequential Autonomic Response:
  • Time-Domain: Stimulation cessation resulted in a significant increase in Mean RR ($-3.18$ ms, $p=0.029$) and decrease in Mean HR ($+6.03$ BPM, $p=0.043$) in chest recordings. AURIS followed these trends but did not reach $p < 0.05$ (likely due to small $n=3$), though effect sizes were high.
  • Complexity Metrics: Showed the most robust responses with large effect sizes:
    • SD1/SD2 Ratio: $d = 1.474$ (Large effect).
    • DFA $\alpha$ Ratio: Significant shift ($p=0.007$) with $d = 1.091$.
• Stability: The system maintained continuous data integrity over multiple trials, resisting motion artifacts that typically degrade thoracic ECGs.

5. Significance and Future Directions

• Clinical Translation: The AURIS platform validates a path toward non-invasive clinical trials for bioelectronic medicine. It eliminates the need for invasive arterial catheters and cumbersome chest electrodes, reducing patient burden and motion artifacts.
• Closed-Loop Potential: By providing reliable, real-time HRV data, AURIS enables the development of closed-loop neuromodulation systems where stimulation parameters are adjusted dynamically based on autonomic tone rather than static protocols.
• Scalability: The successful validation in rodent models suggests the technology is ready for scaling to larger sample sizes and eventual human trials, where the sensor form factor can be customized to individual ear canal anatomy to prevent loosening.

Conclusion: The study successfully demonstrates that the AURIS in-ear sensor is a reliable, gold-standard alternative to chest electrodes for monitoring autonomic tone during spinal cord neuromodulation, overcoming key limitations of current wearable and invasive monitoring technologies.