An inertial-based gait normalcy index for motor fluctuation measurement under an unconstrained walking scenario

The assessment of motor fluctuation is essential for patients with Parkinson's Disease (PD). However, the clinical scales are unable to sufficiently reflect motor fluctuations due to their lower sensitivity to subtle changes. Therefore, in this study, we proposed an inertial-based gait normalcy index (GNI) derived from deviations of gait spatiotemporal parameters related to a healthy age-matched baseline. Eight PD patients were recruited and performed straight-walking and turning tasks under constrained and unconstrained scenarios 4 hours before and after the medication at hourly intervals. The results demonstrated that the GNI could efficiently reflect the motor fluctuation after the medication (p<0.05) and PD patients perform their actual gait ability during the unconstrained straight walking scenarios. The GNI under unconstrained straight walking had a significantly positive high correlation with the UPDRS III score (R = 0.880, p = 0.047). The proposed method has great potential in monitoring motor fluctuation during daily unconstrained activities by providing a more comprehensive biomarker in the future.