In-Ear SpO2: A Tool for Wearable, Unobtrusive Monitoring of Core Blood Oxygen Saturation

Harry J. Davies, Ian Williams, Nicholas S. Peters, Danilo P. MandicSensors 2020 · DOI

TL;DR

This paper demonstrates the feasibility of using an in-ear sensor to accurately measure blood oxygen saturation (SpO2), offering a more convenient and potentially superior alternative to traditional finger-based pulse oximetry. The in-ear approach shows comparable resting SpO2 readings and a significantly faster detection of oxygen level drops, making it promising for continuous, unobtrusive health monitoring, especially in the context of conditions like COVID-19.

What Problem Does This Paper Address?

Accurate and continuous monitoring of blood oxygen saturation (SpO2) is crucial for detecting hypoxemia, a dangerous condition where tissues don't receive enough oxygen. This is particularly relevant for COVID-19 patients, who can experience delayed, life-threatening hypoxemia even without feeling breathless. Traditional pulse oximeters, typically worn on the finger, are often intrusive and can be affected by factors like temperature-induced vasoconstriction. The need for more convenient, unobtrusive, and reliable SpO2 monitoring, especially for long-term use and in wearable health devices ("Hearables"), drives the investigation into alternative measurement sites. This paper specifically addresses the potential of the ear canal as such a site, comparing its performance against the established finger-based method.

Methods

The study employed a custom-designed in-ear photoplethysmography (PPG) sensor incorporating a MAX30101 chip, which emits and detects green, red, and infrared light. This sensor was placed in the right ear canal of participants. Simultaneously, a similar sensor was attached to the right index finger, the conventional measurement site.

The core measurement technique is pulse oximetry, which relies on the differential absorption of red and infrared light by oxygenated and deoxygenated hemoglobin. The ratio of ratios metric was used to calculate SpO2, with the formula:

$R = \frac{AC_{red}/DC_{red}}{AC_{infrared}/DC_{infrared}}$

An empirically derived linear approximation, provided by the manufacturer, was then used to convert this ratio ($R$) into an SpO2 value:

$SpO_2 = 104 - 17R$

To assess the dynamic response, participants underwent a breath-hold protocol. This involved periods of normal breathing, exhaling, and holding their breath, guided by a monitor. The time delay between the drop in SpO2 detected by the ear sensor versus the finger sensor was a key metric. The study also analyzed the amplitude of the PPG signal from both sites.

Datasets

The study utilized data collected from 14 healthy participants (7 males, 7 females) aged 19-38 years. No pre-existing datasets were used; instead, a new dataset was generated through the experimental protocol described above.

Key Insights and Ideas

  • Ear Canal as a Viable SpO2 Site: The ear canal emerges as a promising location for SpO2 measurement, demonstrating comparable accuracy to the finger.
  • Faster Hypoxia Detection: The in-ear sensor detects drops in blood oxygen levels significantly faster than the finger sensor, a critical advantage for early intervention.
  • Robustness to Temperature: The paper alludes to the known robustness of ear canal PPG to temperature-induced vasoconstriction, a significant limitation of finger-based oximetry.
  • Hearables Integration Potential: The findings strongly support the integration of in-ear SpO2 monitoring into "Hearables" (wearable audio devices) for continuous, unobtrusive health tracking.
  • SpO2 Delay as a Health Indicator: The study raises the question of whether SpO2 delay itself could be an indicator of circulatory system health.

Key Results

  • Resting SpO2 Comparison: The root mean square (RMS) difference in resting SpO2 between the ear canal and finger was 1.47%, with a mean difference of 0.23% higher SpO2 in the ear canal. All resting SpO2 values were within the healthy range (94-100%) for both sites, indicating complete agreement on healthy status.
  • SpO2 Delay: The mean relative time delay between the ear canal and finger for detecting minimal blood oxygen was 12.4 seconds. This means the ear canal detected drops in SpO2, on average, 12.4 seconds faster than the finger.
  • Female subjects showed a mean relative delay of 9.67 seconds, while male subjects had a mean of 15.13 seconds (statistically significant difference, P=.045).
  • Absolute delay from the right ear canal was 4.35 seconds (range 0.97-7.31s), while from the right index finger it was 16.75 seconds (range 8.52-28.14s).
  • Photoplethysmogram (PPG) Amplitude: The AC PPG amplitude from the ear canal was significantly lower than from the finger, with the finger PPG amplitude being, on average, 2.35 times higher. However, even with lower amplitudes, the in-ear signals were sufficient for peak detection and SpO2 calculation.

Limitations and Future Work

  • PPG Amplitude: A significant drawback noted is the lower PPG amplitude from the ear canal, attributed to less vascular density and potential sensor contact issues. Future work should focus on improved sensor design, possibly incorporating ear-hooks for better stabilization.
  • Subject Population: The study was conducted on 14 healthy young adults. Future research should investigate SpO2 delay in older cohorts and individuals with underlying cardiovascular conditions.
  • SpO2 Delay as a Biomarker: The paper poses the question of whether SpO2 delay could serve as an indicator of circulatory health, suggesting this as a future research avenue.
  • Left Ear Canal vs. Right Ear Canal: A single subject analysis suggested that the left ear canal might offer an even faster response time than the right, warranting further investigation.

Summary updated

Abstract

The non-invasive estimation of blood oxygen saturation (SpO<sub>2</sub>) by pulse oximetry is of vital importance clinically, from the detection of sleep apnea to the recent ambulatory monitoring of hypoxemia in the delayed post-infective phase of COVID-19. In this proof of concept study, we set out to establish the feasibility of SpO<sub>2</sub> measurement from the ear canal as a convenient site for long term monitoring, and perform a comprehensive comparison with the right index finger-the conventional clinical measurement site. During resting blood oxygen saturation estimation, we found a root mean square difference of 1.47% between the two measurement sites, with a mean difference of 0.23% higher SpO<sub>2</sub> in the right ear canal. Using breath holds, we observe the known phenomena of time delay between central circulation and peripheral circulation with a mean delay between the ear and finger of 12.4 s across all subjects. Furthermore, we document the lower photoplethysmogram amplitude from the ear canal and suggest ways to mitigate this issue. In conjunction with the well-known robustness to temperature induced vasoconstriction, this makes conclusive evidence for in-ear SpO<sub>2</sub> monitoring being both convenient and superior to conventional finger measurement for continuous non-intrusive monitoring in both clinical and everyday-life settings.
BibTeX
@inproceedings{davies2020ear,
  author = {Harry J. Davies and Ian Williams and Nicholas S. Peters and Danilo P. Mandic},
  title = {{In-Ear SpO2: A Tool for Wearable, Unobtrusive Monitoring of Core Blood Oxygen Saturation}},
  booktitle = {Sensors},
  year = {2020},
  doi = {10.3390/s20174879},
}
Figure