Ear worn inertial sensors can estimate scalar kinetic gait parameters and reconstruct vertical ground reaction force curves during running

Ear-worn wearables are widely used while running for entertainment and many modern devices include inertial sensors for user interaction. We propose using earbuds equipped with inertial sensors to estimate the vertical ground reaction forces (vGRFs) from the vibrations transmitted to, and motion of the head. We devised a gait event detection and use this segmentation to train and validate a regression model to predict the vGRFs. Our results showed good agreement with the instrumented treadmill with an average MAPE of 3.97% on scalar GRF derived metrics, showing the promise of ear worn devices for detailed running gait analysis.