Ronny Hannemann
Team Lead ORCA Labs, Erlangen, Germany
Skip to content
The challenge
Unlocking hearing aids for digital health monitoring
Gait is a key indicator of health, with abnormal walking patterns linked to higher risk of falls, cognitive decline, and mental health disorders.
Traditionally, gait tracking requires wearable sensors on the feet, hips, or wrists —devices that can be impractical for continuous daily use.
Modern hearing aids, however, now feature motion sensors, presenting a unique opportunity for unobtrusive, real-time mobility assessment.
Transforming the movement tracking
Until now, no validated gait analysis pipeline existed for ear-worn accelerometers including also spatial-temporal parameters.
This study addresses that gap by developing and validating a system to measure step length, gait speed, and other mobility indicators using hearing aid motion sensors.
If successful, this approach could transform movement tracking, enabling early detection of fall risks and neurological disorders—all without the need for additional devices.
Before our research, no comprehensive gait analysis pipeline existed for ear-worn accelerometers capable of estimating both spatial and temporal gait parameters.
In collaboration with Ann-Kristin Seifer, PhD candidate, as part of a sponsored project by WSA at FAU Erlangen, and Prof. Eskofier’s MaD Lab at FAU Erlangen, we set out to evaluate existing sensor-based gait analysis methods and develop optimized algorithms specifically for hearing aid-integrated motion sensors.
Our approach
Developing a gait analysis pipeline for hearing aids
Our study analyzed walking patterns in younger and older adults at different speeds, using commercial hearing aids with motion sensors alongside gold-standard reference systems such as optical motion capture (OMC) and foot-worn sensors.
To ensure accurate gait assessment, we developed and validated three key components:
- Gait Event Detection – Identifying precise moments of foot contact and toe-off to measure step timing.
- Gait Parameter Estimation – Calculating step length and gait speed using an optimized machine learning model.
- Free-Living Gait Assessment – Evaluating whether hearing aid sensors can reliably track walking patterns in unsupervised, real-world environments.
Rethinking hearing aids as silent guardians of mobility health
Our key insights
Validating hearing aid-based gait analysis
Our analysis confirmed that hearing aid-integrated motion sensors can accurately estimate gait events and temporal parameters, achieving precision comparable to clinical standards.
Additionally, our machine learning model demonstrated robustness across different age groups and sensor settings, though accuracy varied slightly with walking speed.
Key findings include:
- Reliable gait event detection – Foot contact and toe-off moments were accurately identified, matching reference sensor data.
- Accurate step length and gait speed estimation – Motion sensors in hearing aids provided clinically relevant mobility assessments.
- Cross-age validity – The algorithm performed well across younger and older adults, confirming its potential for widespread application.
These results demonstrate that hearing aid motion sensors can serve as practical tools for mobility tracking, offering a real-time, unobtrusive alternative to traditional gait assessment methods.
Future directions
Improving accuracy and clinical applications
These devices successfully detect gait events, estimate walking speed, and track movement patterns across different environments.
To further develop this technology, future research will focus on:
- Expanding clinical applications to assess fall risk and detect early signs of neurological disorders.
- Refining gait analysis algorithms for improved accuracy across various mobility conditions.
- Integrating hearing aid-based monitoring with digital health platforms, allowing healthcare professionals to remotely track patients and deliver personalized care.
Real-world impact
A new era of unobtrusive health monitoring
By leveraging hearing aids for real-time mobility tracking, this research opens new possibilities for fall prevention, mobility management, and early disease detection.
With millions of individuals already wearing hearing aids daily, incorporating health monitoring features offers a scalable, cost-effective solution for both healthcare providers and patients.
The development of the EarGait open-source Python toolbox ensures that researchers and clinicians can refine and expand gait analysis methods, further advancing the field.
As hearing aid technology evolves, these devices could become a mainstream tool for mobility tracking and preventative healthcare.
Close collaboration partners
Researchers
Researchers involved
Research area Benefits beyond hearing
Partners Universities