Appropriate combinations of two IMU sensors achieved mean errors of approximately 30 mm and 11 mm in anterior and mediolateral MoS, respectively, suggesting that IMU-based MoS estimation is sufficiently accurate for the classification of individuals at high fall risk.
Key Findings
Results
Two-IMU combinations achieved mean errors of approximately 30 mm and 11 mm in anterior and mediolateral margin of stability (MoS), respectively, compared with optical motion capture reference values.
IMU sensors were attached to ten body locations while participants walked on a treadmill.
Principal motion analysis, a reductive regression method for multidimensional time-series data, was employed for MoS prediction.
Cross-validation was used for reliable model evaluation.
These errors were described as comparable to the intrinsic standard deviations of MoS.
Results
IMU-based MoS estimation using two sensors was judged sufficiently accurate for classifying individuals at high fall risk.
Mean errors of approximately 30 mm (anterior MoS) and 11 mm (mediolateral MoS) were deemed acceptable.
The achieved errors were comparable to the intrinsic standard deviations of MoS, providing the basis for this sufficiency claim.
The study compared single versus two-IMU configurations to identify optimal sensor placement combinations.
Methods
The study identified specific combinations of two body locations that most effectively predict MoS from IMU signals.
Ten body locations were instrumented with IMU sensors simultaneously.
The study evaluated combinations of two IMU sensor locations for MoS prediction.
Six-axis kinematic signals (linear acceleration and angular velocity) were used as inputs.
Principal motion analysis was the regression method used to relate IMU signals to MoS.
Background
MoS can be approximated from six-axis IMU kinematic signals, extending the applicability of this stability metric beyond optical motion capture laboratory settings.
MoS is typically computed from optical motion capture data.
Previous studies have shown MoS can be approximated from six-axis kinematic signals measured by IMUs.
IMU-equipped devices such as smartphones and smartwatches are increasingly common in daily life.
The current study built on prior single-IMU work by investigating two-IMU combinations.
Peng H, Okamoto S, Watanabe H, Akiyama Y. (2026). Dynamic Gait Stability Estimated Using One or Two Inertial Measurement Units Worn on the Human Body.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26041211