Component-specific kinematic analysis of the timed up and go test using inertial measurement units reflects neuromuscular impairments in patients after stroke.

Abstract

Stroke survivors commonly have impairments in balance, coordination, and motor control that increase fall risk and limit independence. The Timed Up and Go test (TUG) is widely used to assess mobility and fall risk, yet total time may miss neurological deficits. The objective was to determine whether component specific kinematic measures from inertial measurement units provide more sensitive indicators of impairment than a time based outcome. Forty individuals with sub-acute hemiparetic stroke (mean age 73.4 ± 14.6 years; 18 men, 22 women) performed the TUG at maximal speed while wearing inertial measurement units on the pelvis and the feet. The task was segmented into five components: sit to stand, forward gait, turning, return gait, and stand to sit. Neuromuscular impairment was evaluated with the Stroke Impairment Assessment Set (SIAS). Linear regression examined associations between assessment scores and pelvic angular velocities and linear accelerations, adjusting for age and total test time. Higher SIAS scores were associated with greater pelvic roll (B = 1.00, p = 0.003; adjusted R² = 0.48) and yaw (B = 1.88, p = 0.01; adjusted R² = 0.27) angular velocities during the turning phase, even after adjustment for age and TUG time. No significant associations were found for linear acceleration metrics or for total TUG time alone. Phase specific pelvic angular velocities measured by inertial sensors are more sensitive than overall test duration for detecting stroke related motor impairment, particularly during turning. This approach may improve clinical evaluation, support individualized rehabilitation.