Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

RespEar: Earable-Based Robust Respiratory Rate Monitoring

Yang Liu, Kayla-Jade Butkow, Jake Stuchbury-Wass, Adam Pullin, Dong Ma, Cecilia Mascolo

TL;DR

RespEar addresses the challenge of continuous, non-obtrusive RR monitoring across daily activities by leveraging in-ear microphones in earables and exploiting two physiological couplings: RSA and LRC. The system combines RSA-based RR estimation during sedentary periods with LRC-based estimation during active periods, using adaptive signal processing, SSA-based decomposition, and a data-driven pipeline selector to maintain accuracy under noise and motion. Across 18 participants and 8 activities, RespEar achieves an overall MAE of $1.71$ BPM (MAPE $9.68\%$), with $1.48$ BPM ($9.12\%$) sedentary and $2.28$ BPM ($11.04\%$) active errors, outperforming state-of-the-art earable methods. The solution demonstrates real-time viability on a smartphone, robustness to ambient noise, outdoor conditions, and varying speeds, underscoring its practical potential for continuous health monitoring in daily life.

Abstract

Respiratory rate (RR) monitoring is integral to understanding physical and mental health and tracking fitness. Existing studies have demonstrated the feasibility of RR monitoring under specific user conditions (e.g., while remaining still, or while breathing heavily). Yet, performing accurate, continuous and non-obtrusive RR monitoring across diverse daily routines and activities remains challenging. In this work, we present RespEar, an earable-based system for robust RR monitoring. By leveraging the unique properties of in-ear microphones in earbuds, RespEar enables the use of Respiratory Sinus Arrhythmia (RSA) and Locomotor Respiratory Coupling (LRC), physiological couplings between cardiovascular activity, gait and respiration, to indirectly determine RR. This effectively addresses the challenges posed by the almost imperceptible breathing signals under daily activities. We further propose a suite of meticulously crafted signal processing schemes to improve RR estimation accuracy and robustness. With data collected from 18 subjects over 8 activities, RespEar measures RR with a mean absolute error (MAE) of 1.48 breaths per minutes (BPM) and a mean absolute percent error (MAPE) of 9.12% in sedentary conditions, and a MAE of 2.28 BPM and a MAPE of 11.04% in active conditions, respectively, which is unprecedented for a method capable of generalizing across conditions with a single modality.

RespEar: Earable-Based Robust Respiratory Rate Monitoring

TL;DR

RespEar addresses the challenge of continuous, non-obtrusive RR monitoring across daily activities by leveraging in-ear microphones in earables and exploiting two physiological couplings: RSA and LRC. The system combines RSA-based RR estimation during sedentary periods with LRC-based estimation during active periods, using adaptive signal processing, SSA-based decomposition, and a data-driven pipeline selector to maintain accuracy under noise and motion. Across 18 participants and 8 activities, RespEar achieves an overall MAE of BPM (MAPE ), with BPM () sedentary and BPM () active errors, outperforming state-of-the-art earable methods. The solution demonstrates real-time viability on a smartphone, robustness to ambient noise, outdoor conditions, and varying speeds, underscoring its practical potential for continuous health monitoring in daily life.

Abstract

Respiratory rate (RR) monitoring is integral to understanding physical and mental health and tracking fitness. Existing studies have demonstrated the feasibility of RR monitoring under specific user conditions (e.g., while remaining still, or while breathing heavily). Yet, performing accurate, continuous and non-obtrusive RR monitoring across diverse daily routines and activities remains challenging. In this work, we present RespEar, an earable-based system for robust RR monitoring. By leveraging the unique properties of in-ear microphones in earbuds, RespEar enables the use of Respiratory Sinus Arrhythmia (RSA) and Locomotor Respiratory Coupling (LRC), physiological couplings between cardiovascular activity, gait and respiration, to indirectly determine RR. This effectively addresses the challenges posed by the almost imperceptible breathing signals under daily activities. We further propose a suite of meticulously crafted signal processing schemes to improve RR estimation accuracy and robustness. With data collected from 18 subjects over 8 activities, RespEar measures RR with a mean absolute error (MAE) of 1.48 breaths per minutes (BPM) and a mean absolute percent error (MAPE) of 9.12% in sedentary conditions, and a MAE of 2.28 BPM and a MAPE of 11.04% in active conditions, respectively, which is unprecedented for a method capable of generalizing across conditions with a single modality.
Paper Structure (33 sections, 2 equations, 20 figures, 1 algorithm)

This paper contains 33 sections, 2 equations, 20 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. Preliminary Investigation
  3. Initial Exploration: Sensors on Earables
  4. Design Primer
  5. RSA-based RR Estimation
  6. LRC-based RR Estimation
  7. System Design
  8. RSA-based RR Monitoring
  9. Design Principle
  10. Our design.
  11. HRV Signal Estimation.
  12. Adaptive Breathing Signal Extraction.
  13. Interference Artefact Filtering.
  14. LRC-based RR Monitoring
  15. Stride Frequency Estimation.
  16. ...and 18 more sections

Figures (20)

  • Figure 1: RespEar offers RR monitoring across a variety of daily activities using earbuds.
  • Figure 2: Signals captured using the (a) IMU, (b) out-ear and (c) in-ear microphone under different activities.
  • Figure 3: Illustration of the RespEar architecture.
  • Figure 4: (a) Distribution of $RR_{GT}$ while sedentary. (b) Comparison of RR estimation performance using different BPFs.
  • Figure 5: (a) In-ear audio and heartbeats. (b) Peak detection.
  • ...and 15 more figures