Validity and reliability of inertial measurement units for measuring gait kinematics in older adults across varying fall risk levels and walking speeds

BackgroundThis study aimed to determine whether the validity and reliability of inertial measurement units (IMUs) in measuring lower extremity joint kinematics differed among different walking velocities and varying fall risk older adults.MethodsForty-five older adults were categorized into low and high fall risk. Lower extremity joint angles during walking were simultaneously measured using inertial measurement units (IMUs) and an optical motion capture (OMC) system across slow, preferred, and fast walking speeds. The Coefficient of Multiple Correlation (CMC) was employed to assess waveform consistency, while systematic error (SE) was calculated to quantify deviations. For discrete kinematic parameters, Pearson's correlation coefficient (r), intra-class correlation coefficients (ICCs), and root mean square error (RMSE) were computed. A two-way analysis of variance utilizing statistical parametric mapping methods was conducted to compare differences in lower extremity joint angles across fall risk categories and varying walking speeds.ResultsFor lower extremity joint waveforms, IMUs demonstrated very good to excellent validity (low fall risk: CMC = 0.872-0.957, SE = 4.8 degrees-9.6 degrees; high fall risk: CMC = 0.894-0.974,SE = 4.5 degrees-6.9 degrees) and reliability (within-raters: low fall risk: CMC = 0.924-0.982, SE = 3.9 degrees-6.5 degrees; high fall risk: CMC = 0.914-0.974, SE = 1.0 degrees-2.5 degrees; between-raters: low fall risk: CMC = 0.953-0.981, SE = 2.2 degrees-5.9 degrees; high fall risk: CMC = 0.903-0.985, SE = 1.7 degrees-2.2 degrees) for joint kinematics in the sagittal plane across various walking velocities in both low and high fall risk individuals. For discrete parameters, r values of peak joint angles and ROM measured by IMUs and OMC system were generally higher in the sagittal plane compared to the frontal and transverse planes. ICC values between within-raters and between-raters comparisons ranged from moderate to excellent, with RMSE values between 2.1 degrees- 8.4 degrees for both within-rater and between-rater comparisons in the sagittal plane. There are significant differences in the joint angles of lower limbs between different fall risk groups and different walking speeds in specific gait cycle intervals.ConclusionIMUs are valid and reliable for measuring joint kinematics both low- and high-fall risk older adults across different walking speeds. However, caution is warranted when interpreting data from the frontal and transverse planes. Additionally, high fall risk individuals had smaller joint angles, while faster walking speeds increased joint angles during specific gait cycles.